Open Access Article

The authors, the publisher, and the right holders grant the right to use, reproduce, and disseminate the work in digital form to all users.

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 2, (2011) Adenine Press (2011)

Vocabulary of Denitions of Life Suggests a Denition

http://www.jbsdonline.com

Edward N. Trifonov1,2
1Genome

Diversity Center, Institute of

Evolution, University of Haifa, Mount

Abstract Analysis of the vocabulary of 123 tabulated denitions of life reveals nine groups of dening terms (denientia) of which the groups (self-)reproduction and evolution (variation) appear as the minimal set for a concise and inclusive denition: Life is self-reproduction with variations. Key words: Consensus; Denientia; Evolution; Origin of life; Self-reproduction; Variations; Vocabulary.

Carmel, Haifa 31905, Israel

2Department

of Functional

Genomics and Proteomics, Faculty of Science, Masaryk University, Kotlarska 2, CZ-61137 Brno, Czech Republic

Over 100 of denitions of life exist today (1, 2) learned opinions each one of which is, or has been in the past, defended not without a reason though generally met with skepticism. The skepticism is multiplied by the above number, leaving almost no chance for new formulations which, however, continue to appear. An excellent overview of the current status of the problem is given by a special issue of the journal Origins of Life and Evolution of Biospheres (3). Sixteen papers of this issue, expert opinions, display a variety of philosophical and historical aspects of dening life, and inevitable limitations of about every approach and view point. The denitions are more than often in conict with one another. Undeniably, however, most of them do have a point, one or another or several, and common sense suggests that, probably, one could arrive to a consensus, if only the authors, some two centuries apart from one another, could be brought together. One thing, however, can be done sort of voting in absentia asking which terms in the denitions are the most frequent and, thus, perhaps, reecting the most important points shared by many. Such analysis is offered below, revealing those most frequent terms that may be used for tentative formulation of the consensus. In the Table I the vocabulary of words used in 60-denition set of Barbieri (1) and 90-denition collection of Popa (2) is presented. The non-redundant total size of two collections is 123 denitions. All words of 3 or more letters are taken for the survey, excluding connective ones (the, and, that, etc.). The words that appear more than 4 times in the collections are presented in the Table I (the full list is available by request). The life, as deniendum, is at the top of the list. Inspection of the list reveals that amongst frequent words the ones closely related to, e.g., life group (such as living, alive) appear as well. This suggests combination of various words in groups by their common meaning. The Table II displays several such groups, topmost by their scores. Words of each group are present in at least 30% of the denitions analyzed. The groups of smaller size (not shown) contain, essentially, only words with the same root (e.g., denition, dened, dening, etc.).

Phone: +972 4 828 8096 Fax: +972 4 824 6554 E-mail: trifonov@research.haifa.ac.il

259

260 Trifonov

Thus, the consensus of the life definition patched from these nine definientia would be: Life is [System, Matter, Chemical (Metabolism), Complexity (Information), (Self-)Reproduction, Evolution (Change), Environment, Energy, Ability,] where the square brackets correspond to some compact expression containing the words listed within. For example, one possibility is:

Life is metabolizing material informational system with ability of self-reproduction with changes (evolution), which requires energy and suitable environment.

[1]

Since the analysis described, inevitably, is not free from some arbitrariness in assignment of the words to this or another group, it would be desirable to compare it to a similar study conducted by an independent laboratory. The work of Kompanichenko (4) offers such independent analysis. In that work the definitions of life are taken from the collection of papers and definitions (5) that preceded the publication of Popa (2). Kompanichenko considered definitions given by 63 authors from the book of Palyi et al. (5), and extracted from the definitions 19 major unique fundamental properties of biological systems. Instead of words the notions of the properties have been used, often expressed by several words, differently by different authors. The largest category, according to Kompanichenko, is Capable of evolution including the increase of complexity, hierarchy and the display of self-perfecting logic (close to Evolution in our list) 32 authors. The fundamental property Capability for self-reproduction (Reproduction in our list) has been mentioned by 27 authors. Another fundamental property, Capability for self-replication (11 authors) also would go to Reproduction group, 38 authors together. Note that according to the notion count of Kompanichenko our categories Evolution and Reproduction appear at the top of the list, while by word count (this work) they rank more modestly. Then follow Performance and control of metabolism, including autocatalysis, cyclic chemical processes, feedback loops, and active transport (Chemical of our list) 25 authors, Ability to extract (free) energy and matter from the environment (Energy, Matter, Environment, and Ability in our list) 16 authors, and Capacity to accumulate, re-organize and transmit genetic information (Complexity/Information) in our list) 13 authors. Remaining 13 properties considered by Kompanichenko are mentioned by smaller number of authors (1 to 7) and are not reflected in our list of major terms. This approach, obviously, is not free from subjective assignments as well. It is remarkable, however, that eight of nine categories of our list of major terms are also at the top of the list of Kompanichenko.
Table I List of most frequent words in the definitions of life. Life Living System Matter Systems Environment Energy Chemical Process Metabolism Organism Organization Complexity Ability Itself Able Capable Definition 123 47 43 25 22 20 18 17 15 14 14 14 13 12 12 11 11 11 Organic Alive Evolution Materials Reproduction Existence Defined Growth Information Open Processes Properties Property Reproduce Through Complex Evolve Genetic 11 10 10 10 10 9 8 8 8 8 8 8 8 8 8 7 7 7 Internal Replication Being Change Characteristics Entity External Means Molecules One Order Organisms State Things Time Way Based Biological 7 7 6 6 6 6 6 6 6 6 6 6 6 6 6 6 5 5 Capacity Different Force Form Functional Highly More Mutation Necessary Network Objects Only Organized Reactions Self-reproduction Some Three 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5

Table II Groups of words with similar meaning. LIFE living alive being biological other related words Sum SYSTEM systems organization organism order organisms network organized other related words Sum MATTER organic materials molecules other related words Sum CHEMICAL process metabolism processes reactions other related words Sum 123 47 10 6 5 8 199 43 22 14 14 6 6 5 5 40 155 25 11 10 6 36 88 17 15 14 8 5 26 85 COMPLEXITY information complex other related words Sum REPRODUCTION reproduce replication self-reproduction other related words Sum EVOLUTION evolve change mutation other related words Sum ENVIRONMENT external other related words Sum ENERGY force other related words Sum ABILITY able capable capacity other related words Sum 13 8 7 46 74 10 8 7 5 33 63 10 7 6 5 20 48 20 6 15 41 18 5 17 40 12 11 11 5 1 40

261 Definition of Life

Although the extract [1] above may already appear as a reasonable definition, one, of course, would like to have it more concise, and desirably, containing components that are both necessary and sufficient, either alone or in combination. The possible shorter definitions would be, of course, subject of a thorough evaluation, inviting new discussions of the definition problem, but on a new basis of limited number of relevant terms. Below one rather plausible reduction is suggested. First, some of these consensus terms implicitly involve others. For example, existence of metabolism implies both energy and material supply which also represent environment. Self-reproduction (replication) appears to be the most inclusive term of the nine groups above, as it implies metabolism and system as well. That is, if the selfreproduction is going on, it can proceed only on condition that metabolism, system, energy and material supply are also in place. The complexity (information) can be considered also as product of self-reproduction with changes (evolution), on the evolutionary route from simple to complex. We are, thus, left with two independent notions: self-reproduction and changes (evolution). None of these two implies another one. They, actually, exclude one another, as self-reproduction is exact copying, no changes, while changes can not relate to exact copying. These two notions can be combined in a third one: an almost exact self-reproduction or self-reproduction with variations, suggesting, thus, a tentative minimalistic definition. Evolution (and natural selection) means changing inheritance, that is, causing variations in self-reproduction. As it has been said by Darwin (6): if variations useful to any organic being ever do occur, assuredly individuals thus characterized will have the best chance of being preserved in the struggle for life; and from the strong principle of inheritance, these will tend to produce offspring

262 Trifonov

similarly characterized (6, Chapter 4, italics by ENT). The definition of life based on only two terms extracted from the vocabulary of definitions, and consistent with Darwins views would be, thus:

Life is self-reproduction with variations.

[2]

Here the tandem self-reproduction with variations should be considered as one indivisible term of very clear Darwinian meaning. Of 123 different definitions only 18 contain this pair of definientia, in combination with other defining terms of the Table II. The most succinct among them is one by Oparin (7):

Any system capable of replication and mutation is alive.

The vocabulary approach implemented in this work is conceptually close to the Principal Component Analysis (8, 9) which is applicable to large ensembles of quantitative data. The data are reduced to several independent (orthogonal) components, and the major principal component is extracted that covers most of the data. Similarly, our vocabulary of definitions is reduced to several groups of words with different meaning. The same procedure of extraction of a single principal component has been used in derivation of consensus temporal order of engagement of amino acids in early evolution (10, 11). The order has been established as a consensus of a large number of rather different chronologies suggested by various authors. As this order correlates well with thermostability of respective codonanticodon pairs, the reconstruction of the Evolutionary Chart of Codons became possible. From the properties of the Chart several important features of the earliest stages of molecular evolution have been predicted, and confirmed by sequence analyses (12, 13). The earliest steps of the evolution of the codons also suggested two major stages in the origin of life self-reproduction (exact replication of the ideal RNA duplex in the above theory, one strand of which is repeating triplet GCCn, while another strand is complementary GGCn), and variations (appearance of pointmutated versions of GCC and GGC in the subsequent replications). That ended logically in the definition of life self-reproduction with variations (14), identical to above [2]. In its earlier version almost precise replication it appeared in the collection of definitions of life gathered by Barbieri (1). (That formula has been excluded from the analysis above and did not enter the vocabulary). According to this model-based definition, any experimental work involving the GCCn* GGCn replicator would border the life-nonlife transition. The definition of life, thus, is naturally required for the exploration, at least as a practical guide in the research (15). Thus, it is not purely philosophical and historical matter anymore. This was one of the motivations for the linguistic analysis described in the paper. The derived definition [2] is minimalistic both by definientia involved and, independently, by structurally minimal size of the presumed earliest replicator to which the definition fully applies (14). One unforeseen property of the minimalistic definition is its generality. It can be considered as applicable not just to earthly life but to any forms of life imagination may offer, like extraterrestrial life, alternative chemistry forms, computer models, and abstract forms. It suggests a unique common basis for the variety of lives: all is life that copies itself and changes. One important question to address: is the minimalistic definition both necessary and sufficient? Even most primitive forms of observable life are still too complex, to claim that they can be reduced to the above simple formula. The applicability of the definition can only be tested on much simpler artificial life-like models which one day will, hopefully, be designed and brought to life, by providing artificially

produced necessary ingredients. This day is not far away. A self-catalytic generation (cross-replication) of plus-strand and minus-strand ribozymes from their constituent 14-mer and 52-mer RNA chains has been recently accomplished (16). The system also allows to introduce and to monitor evolutionary changes in the ribozymes. It does not include, though, the elementary template polymerization steps one would expect the simplest self-reproducing system to have. The experimental chemical template polymerization has been intensively studied during last two decades (for most recent review see (17)). An efficient complementary primer extension on C15 template has been achieved in protocells by using chemical RNA analog (18). No evolutionary changes have been observed so far in the system. Yet simpler setup, with oligo-riboA in aqueous solution has been developed, in which the chain of RNA could elongate indefinitely, apparently, due to formation of complementary contacts between polyA chains (19). In this work similar extension of oligoG ligated to oligoC has been observed as well, with incorporation of complementary Gs. Creation of a bacterial cell with chemically synthesized genome (20) has to be mentioned as another case of a system at the border lifenon-life. This is, actually, a very large scale bacterial transformation where the transforming DNA has been, indeed, chemically synthesized according to natural design previously fully sequenced genome, with some changes. The synthesized genome did replicate many rounds. However, it should be considered as very much assisted replication as it was provided with an initial natural cytoplasm. No evolutionary changes in the design have been monitored. Nearly fully artificial life with the properties described by the above minimalistic definition has been created many years earlier, by Sol Spiegelman his famous monster of replicating degenerate products of mutating Q-beta RNA. The monster RNA versions have practically lost any sequence similarity to their viral ancestor, in the process of ingenious evolutionary game (21). This has been an assisted replication as well, since the experiments have been performed in presence of natural replicase. However, this protocol is, probably, the most promising starting model for eventual design of simplest truly artificial life, since for modern peptide chemistry the synthesis of an active analog of the replicase, perhaps, is as plausible as the chemically synthesized genome. Acknowledgements Author is grateful to anonymous reviewers for thoughtful comments, criticism and suggestions towards improvement of the paper.
References 1. M. Barbieri. The Organic Codes. An introduction to Semantic Biology. Cambridge University Press, Cambridge (2003). 2. R. Popa. In Between Necessity and Probability: Searching for the Definition and Origin of Life. Series: Adv Astrobiol Biogeophys, Springer, NY, pp. 197-205 (2004). 3. J. Gayon, C. Malaterre, M. Morange, F. Raulin-Cerceau, and S. Tirard (guest Eds.). Special Issue: Definitions of life. Origins Life Evol Biospheres 40, 119-244 (2010). 4. V. N. Kompanichenko. Int J Astrobiol 7, 27-46 (2008). 5. G. Palyi, C. Zucci, and L. Caglioti (Eds.). Fundamentals of Life. Elsevier SAS, Paris (2002). 6. C. Darwin. Origin of species, John Murray, London (1859). 7. A. I. Oparin. as referred to in (2) (1961). 8. K. Pearson. Philosophical Magazine 2, 559-572 (1901). 9. I. T. Jolliffe. Principal Component Analysis. Springer Series in Statistics, 2nd ed., Springer, NY (2002). 10. E. N. Trifonov. Gene 261, 139-151 (2000). 11. E. N. Trifonov. J Biomol Struct Dyn 22, 1-11 (2004). 12. E. N. Trifonov. Isr J Ecol Evol 52, 375-387 (2006). 13. E. N. Trifonov. J Cosmology 10, 3374-3380 (2010). 14. E. N. Trifonov. Res Microbiol 160, 481-486 (2009). 15. S. Pino, E. N. Trifonov, and E. Di Mauro. Genomics, Proteomics and Bioinformatics 9, 7-14 (2011). 16. T. A. Lincoln and G. F. Joyce. Science 323, 1229-1232 (2009).

263 Definition of Life

264 Trifonov

17. J. S. Schrum, T. F. Zhu, and J. W. Szostak. Cold Spring Harb Perspect Biol 2, a002212 (2010). 18. S. S. Mansy, J. P. Schrum, M. Krishnamurthy, S. Tob, D. A. Treco, and J. W. Szostak. Nature 454, 122-125 (2008). 19. G. Costanzo, S. Pino, F. Ciciriello, and E. Di Mauro. J Biol Chem 284, 33206-33216 (2009). 20. D. G. Gibson, J. I. Glass, C. Lartigue, V. N. Noskov, R.-Y. Chuang, M. A. Algire, G. A. Benders, M. G. Montague, L. Ma, M. M. Moodie, C. Merryman, S. Vashee, R. Krishnakumar, N. Assad-Garcia, C. Andrews-Pfannkoch, E. A. Denisova, L. Young, Z.-Q. Qi, T. H. Segall-Shapiro, C. H. Calvey, P. P. Parmar, C. A. Hutchison III, H. O. Smith, and J. C. Venter. Science 329, 52-56 (2010). 21. D. R. Mills, F. R. Kramer, C. Dobkin, T. Nishihara, and S. Spiegelman. Proc Natl Acad Sci USA 72, 4252-4256 (1975).

Date Received: March 17, 2011

Communicated by the Editor Ramaswamy H. Sarma

Appendix Full list of the words making groups of the Table II. LIFE living alive being biological animate beings animated Sum SYSTEM systems organization organism order organisms network organized assembly building components composed ensemble aggregates automata consists ordered arrangement automaton automatons biosystem built consisting contain containing ensembles multilevel networks orderly organizational organize rearrangement self-organization self-organized Sum MATTER organic materials molecules material compounds nucleic polymers proteins acids fluid substance acid aqueous bioelements carbon carbon-based molecule monomers oligosaccharides 123 47 10 6 5 4 3 1 199 43 22 14 14 6 6 5 5 3 3 3 3 3 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 155 25 11 10 6 4 3 3 3 2 2 2 2 1 1 1 1 1 1 1 1 polymer polypeptides polysaccharides protein proteinaceous substances water Sum CHEMICAL process metabolism processes reactions molecular production metabolic metabolize chemistry produces decay degradable degradation exchange precursors processing produced regeneration reparation synthesis synthesize Sum COMPLEXITY information complex code entropy knowledge patterns communicate molecular-informational pattern program reading algorithmic complicated computational digital feedback feedbacks feedback-loops informational informationally informationally-controlled information-storage instructional instructions low-entropy maximally-complex message program-controlled programs self-correction self-instruction self-reading 1 1 1 1 1 1 1 88 17 15 14 8 5 4 3 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 85 13 8 7 4 3 3 3 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

265 Definition of Life

(Continued)

266 Trifonov
sequence signal Sum REPRODUCTION reproduce replication self-reproduction self-replication autopoiesis autopoietic multiplication proliferation replicate self-replicating self-reproduce self-reproducing copies copying perpetuate perpetuated proliferating recreate reproduces reproducibility reproducing self-duplication self-generating self-generation self-perpetuating self-producing Sum EVOLUTION evolve change mutation changes evolutionary mutate variation errors evolved evolves 1 1 74 10 8 7 5 3 2 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 63 10 7 6 5 4 2 2 2 2 1 1

Appendix (Continued) mutability modifies mutandis mutations mutatis variant Sum ENVIRONMENT external conditions surroundings biosphere condition conditionally conditioned environmental environments medium microenvironment supply Sum ENERGY force thermodynamic engine forces power powers energetically energy-dependent energies engines thermodynamical thermodynamics Sum ABILITY able capable capacity capacities Sum 1 1 1 1 1 1 48 20 6 3 3 1 1 1 1 1 1 1 1 1 41 18 5 3 2 2 2 2 1 1 1 1 1 1 40 12 11 11 5 1 40

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Editorial

A Conversation on Definition of Life

http://www.jbsdonline.com In his letter of submission of the Definition of Life paper (1), Edward Trifonov indicated
The subject is very much debatable, the analysis may appear to some controversial, and the claim outrageous. That also means, that the paper could be suggested for open discussion. I am ready to confront any challenges.

Ramaswamy H. Sarma
Department of Chemistry, State University of New York at Albany, Albany NY 12222, USA

Well, I thought, let me see what the referees have to say. One referee in part said:
I happen to disagree with Edward Trifonov, and yet I strongly recommend the publication of his paper. The reason is that the definition of life is an extremely important issue but also one where there is virtually no objective approach. Trifonovs paper is probably the first of this kind, and it is for that reason that it should be circulated. When it is published, it shall be possible to discuss it and eventually propose something better, but there must be a starting point.

Interesting, this referee disagrees with the author, but wants me to publish the paper to start a Conversation among his peers. Wow! What a magnanimous gesture! And a second referee in part said:
And indeed, the resulting definition thus laboriously refined from all the previous ones strikes the reader as very concise and straightforward. I must confess that my first reaction to this move by the author was how come nobody thought about it so far? All in all, this paper combines a brilliant and rigorous analysis with a profound scientific and philosophical result. It certainly deserves being published as is in any leading scientific journal.

Then we heard from referees about the absence of reproducibility, controls, and the utter lack of the consideration of central elements in life such as beauty, truth and love, a definition of life devoid of life and soul, manufactured from linguistics. I thought that the effort by Trifonov was an interesting intellectual adventure that warranted open comments and discussion by scientists and philosophers researching in life, its origins (2-6) and evolution (7-11). So I extended them an invitation and most of them agreed to provide the comments. Prof. Pier Luigi Luisi did not write a formal response to Trifonovs paper, but wrote me the letter reproduced verbatim below:
Dear Rama I have been thinking about it, and definitely decided not to invest my time to write an article, as it would be to give relevance to what I consider a poor piece
Corresponding author: Ramaswamy H. Sarma Phone: 518-456-9362 Fax: 518-452-4955 E-mail: rhs07@albany.edu

597

598
of science-and I do not want to do that. I would like however to briefly explain to you the reasons for my negative stand. What Trifonov does, is to make a kind of average of all possible life definitions of life, giving to everyone the same statistical weight. A democratic decision. But science does not work this way. It comes to my mind the statistic that some politicians do: given a group of scientists who says: two and two makes four; and another group who say: two and two make sixthe politicians say, well, let us not argue, lets make so that two and two makes five... This kind of statistics is simply wrong, non-sense things should first be eliminated. Of course this necessitates an arbitrary act of courage-this is the responsibility of the scientist. And then the basic result of our Author, that life is reproduction and variation-which means, change, evolution... my God, we have been debating for so many years that this is not simply so. The old grand mother of our author is not capable of reproduction, but is (I hope) still living, and so are all women of this planet over 60 years. Not able to reproduce...and then not living? And to decide whether an oak tree is living, you wait a few hundred years until it reproduces? Reproduction is important, but it is a consequence of life, it can be there or not, it depends...this is so obvious, and common sense is much more important than statistics. And the confusion with variation-changes, evolution: a colony of bacteria which is not reproducing in a measurable time scale-is not living? and again, to decide whether something is living you wait until you measure changes? which kind of? As I told you, Rama, these issues were already debated in my old paper (1998) on the definition of life and in my book on the emergence of life-long ago. Trifonov does not mention all this, and he is right, in the sense that I am now completely on another place; let me only add that, when I read a paper, I always ask myself whether this is something I would suggest to my students. In this case, my answer is definitely negative, as they would not learn anything-they would probably get more confused. Dear Rama, although my decision is definitive, if you want/need to use some of the material in this letter, you can do it, thanks for the trust. Luigi
References P. L. Luisi. About various definitions of life. Origins of Life and Evolution of the Biosphere 28, 613-622 (1998).

Sarma
P. L. Luisi. The Emergence of Life: From Chemical Origins to Synthetic Biology, Cambridge Univ. Press, Cambridge UK (2006).

In my letter of invitation, I made it explicitly clear that their comments would not be subject to the normal peer review so that they could express their personal points, views, and evaluation of Trifonovs paper without interference from the referees. This is perfectly fine because the comments themselves are not research articles which require mandatory peer evaluations, but just pure comments, and referees inserting moderation and balance into the comments is not right. These comments are brief items with a short list of references, and do not contain original research data. They are essentially open referee reports. The comments are published without dates received and without the name of the Communicating Editor because they are not regular research articles. Finally this Journal is publishing the section consisting of this editorial, the 19 comments and the author response as Open Access. This is because this Journal strongly believes that doctoral students in biochemistry and molecular biology will benefit a great deal from a study of these comments; and Open Access publication enables this. I have received comments from 19 laboratories across the globe. I thank all of them for reading Trifonovs paper and expressing their opinion. I am particularly grateful to Nobel Laureate Jack Szostak for visiting Albany, delivering the Keynote address at the 17th Conversation, and chairing the session on origin of life there and for participating in this Conversation.
References 1. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). 2. J. W. Szostak. J Biomol Struct Dyn 28, 1059-1059 (2011). 3. U. J. Meierhenrich, J.-J. Filippi. C. Meinert, J. H. Bredehft, J.-i. Takahashi, L. Nahon, N. C. Jones, and S. V. Hoffmann. J Biomol Struct Dyn 28, 1060-1061 (2011). 4. G. Costanzo, S. Pino, F. Ciciriello, and E. Di Mauro. J Biomol Struct Dyn 28, 1061-1061 (2011). 5. J. P. Ferris, P. C. Joshi, M. F. Aldersley, and J. W. Delano. J Biomol Struct Dyn 28, 1062-1062 (2011). 6. A. E. Engelhart, R. Buckley, E. D. Horowitz, and N. V. Hud. J Biomol Struct Dyn 28, 1063-1063 (2011). 7. W. L. Duax, R. Huether, D. Dziak, and C. McEachon. J Biomol Struct Dyn 28, 1066-1067 (2011). 8. A. Y. Panchin, E. N. Shustrova, and I. I. Artamonova. J Biomol Struct Dyn 28, 1068-1068 (2011). 9. L. D. Williams, C. Hsiao, J. C. Bowman, C. R. Bernier, J. Peters, D. M. Schneider, and E. ONeill. J Biomol Struct Dyn 28, 1071-1072 (2011). 10. T. J. Glembo and S. B. Ozkan. J Biomol Struct Dyn 28, 1068-1069 (2011). 11. A. Goncearenco and I. N. Berezovsky. J Biomol Struct Dyn 28, 1065-1065 (2011).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

Attempts to Define Life Do Not Help to Understand the Origin of Life

http://www.jbsdonline.com Attempts to define life are irrelevant to scientific efforts to understand the origin of life. Why is this? Simply put, the study of the origin of life is an effort to understand the transition from chemistry to biology. This fundamental transition was the result of a lengthy pathway consisting of many stages, each of which is the subject of numerous scientific questions. Simple chemistry in diverse environments on the early earth led to the emergence of ever more complex chemistry and ultimately to the synthesis of the critical biological building blocks. At some point, the assembly of these materials into primitive cells enabled the emergence of Darwinian evolutionary behavior, followed by the gradual evolution of more complex life forms leading to modern life. Somewhere in this grand process, this series of transitions from the clearly physical and chemical to the clearly biological, it is tempting to draw a line that divides the non-living from the living. But the location of any such dividing line is arbitrary, and there is no agreement on where it should be drawn. An inordinate amount of effort has been spent over the decades in futile attempts to define life often and indeed usually biased by the research focus of the person doing the defining. As a result, people who study different aspects of physics, chemistry and biology will draw the line between life and non-life at different positions. Some will say there is no life until a well defined set of metabolic reactions are in place. Others will focus on spatial compartmentalization, on the various requirements for Darwinian evolution, or on the specific molecules of inheritance. None of this matters, however, in terms of the fundamental scientific questions concerning the transitions leading from chemistry to biology the true unknowns and subject of origin-of-life studies. Beyond the arbitrary nature of efforts to define the boundary between non-life and life, this effort is illusory for a deeper reason. As one focuses experimentally on any of the defining properties of life, the sharp boundary seems to blur, splitting into finer and finer sub-divisions. As an example, let us look at the emergence of Darwinian evolution, which is often cited [e.g., see Table II, in ref. 1] as a key aspect of the definition of life (with good reason, as Darwinian evolution is indeed the unifying characteristic of all of biology). Certainly once cells with genetically encoded advantageous functions existed, classically defined Darwinian evolution had begun, and most people would define such cells as alive. But what about the previous steps? Such cells would likely have been preceded by protocells, with replicating genetic information, but lacking coded functions that provided a cellular advantage. At this stage, replication with heritable variation would have existed, and whatever process drove replication would most likely have had biases that led to changes in the genetic structure of the population over time. Would that minimalist form of evolution qualify such protocells as being alive? Going back even further, consider genetic molecules replicating in solution

Jack W. Szostak
Howard Hughes Medical Institute and Department of Molecular Biology and Center for Computational and Integrative Biology, Massachusetts General Hospital, 185 Cambridge Street, Boston, Massachusetts 02114

Corresponding author: Jack W. Szostak E-mail: szostak@molbio.mgh.harvard.edu

599

600
or on particulate surfaces again, biases in replication would lead to selection for sequences that are better templates, i.e., easier to replicate. Even the assembly of the first genetic polymers would have had biases, leading to non-random population structures. Darwinian evolution itself emerged in a series of stages, step-by-step, gradually leading to the almost infinite potential for organismal variation seen in modern biology. And yet, to define a single point along the progression as the point at which Darwinian evolution first emerged would be difficult. More importantly, such a definition would not further our understanding of the transitions involved or the nature of the physical and chemical forces driving those transitions.

Szostak
What is important in the origin of life field is understanding the transitions that led from chemistry to biology. So far, I have not seen that efforts to define life have contributed at all to that understanding. Acknowledgement I thank Noam Prywes and Aaron Englehart for helpful comments on this manuscript.
Reference 1. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

Trifonovs Meta-Definition of Life

http://www.jbsdonline.com What is a Definition in the Realm of Biology? The problem of definition is at the basis of the problem of understanding. Any level of understanding of any natural phenomenon or object entails its description, followed by the comparison (in our brain or in the pages of an Herbarium, or by a computer program) with similar phenomena or objects. Between comparison and understanding we encounter classification. The problem of classification is not solved. The tradition we are acquainted with is the Linnaean System. Simplifying (too much, I am afraid), according to this system every organism belongs to a Species and to a Genus. The properties allowing the organism to be encased in a Species are, in semiological terms, Dictionarian (that is: context-free). The characters for the Genus are Encyclopaedian (context-dependent). Given that the two categories are based on principles that do not belong to the same category, in terms of logics the System is ambiguous. Linnaean classification has had some use essentially because it empirically approximates the definiendum to the closest functional happenstance. The upper floors of the System (Families, etc.) suffer from the same approximation. Similar ambiguities characterize other Systems. Aristotelian classification was, in spite of its almost bi-millennial life, even more logically ill-based. To the point that when a careful analysis was made, it was recognized by Porfirius that applying its principles with some rigour one would quickly hit the limit set by penuria nominum, scarcity of names, impossibility to classify providing the appropriate labels. The era of genomics has clarified this point with precision. Each genome is itself: similarities abound, as differences do. The basic structural, functional and informational principles of living entities are the same since the very beginning of their history. These principles essentially consist of the organization of a genotype (the egg) gathering, maintaining and transmitting information related to itself and to a phenotype (the chicken), harnessing and directing energy and matter into the further organization, maintenance and transmission of the egg. From the earliest information to the extant enormous (and informationally limitless) genomes no interruption exists (by definition), no firm border can be traced. Shuffling genes from one genome to others, constructing genetic chimaeras made of genes originated in different Kingdoms or, simply, just performing meta-genomics of flasks of sea water shows the principle of penuria nominum in contemporary terms. In the realm of biology there are no barriers, definitions are elusive. Thats where Trifonovs thought comes into play.

Ernesto Di Mauro
Dept. of Biology and Biotechnologies Charles Darwin University Sapienza, P.le Aldo Moro, 5, 00185 Roma, Italy

Corresponding author: Ernesto Di Mauro Phone: +39.06.4991.2880 Fax: +39.06.49912500 E-mail: ernesto.dimauro@uniroma1.it

601

602
A Radical Approach The contribution of this paper is in its hidden radical criticism, and in the solution it proposes. My personal opinion is that no acceptable definition of life exists, yet. According to the most popular one, life is a selfsustained chemical system capable of undergoing Darwinian evolution (2). This is very close to the definition provided by Oparin: Any system capable of replication and mutation is alive (3). However, life is a process, not a system. In addition, if a definition relies on the variation (evolution) of its definiendum, it is intrinsically a description, more than a definition. This does not diminish its empirical value, in a sense very close to what we have mentioned about the purport of the Linnaean classification System. Rather, these two definitions help to make the point: we are dealing with descriptions, not definitions. Trifonovs title implies exactly this: all definitions are relative. Corollary: a relative definition is not a definition. In a given frame of reference a law is absolute, or is not. Comparing the proposed definitions, as he does, is extremely useful (and original). Especially so if done with the rigour and the wideangle that characterize his Gnomic (4) approach. What turns out is that comparing the definition distilled from his reported tabulation of the 123 definitions analyzed (namely: life is self-reproduction with variations) with the currently most accepted life is a self-sustained chemical system capable of undergoing Darwinian evolution, the only common term is self . Which may well be the final minimalist definition of life, encompassing them all. A Way Out from Ambiguity The difference between description and definition is not just semantics, is categorical. May the two categories be reconciled? Trifonovs reasoning provides a solution: at the cross between the two lays, in the realm of biology, Darwins

Di Mauro
warm little pond (5). There is where it is believed that everything started, where the first molecules began accumulating, replicating and evolving information. His three decades-long Gnomic approach consists of the compilation and analytical comparison of essentially all that is known about sequences (nucleic acids, proteins and, in between, coding functions) looking for the very first principles. Gnomic is description aiming to definition. I believe that he has come very close to the solution: The earliest steps of the evolution of the codons also suggested two major stages in the origin of life self-reproduction (exact replication of the ideal RNA duplex in the above theory, one strand of which is repeating triplet GCCn, while another strand is complementary GGCn), and variations (appearance of point-mutated versions of GCC and GGC in the subsequent replications). (1, and references therein). Accordingly, studies involving the GCCn*GGCn replicator border the life-non life transition (6), reducing it to initial experimental analysis (7). The identification of plausible first replicators would help to pinpoint the events that lead from dis-order to order, and ignite the process that we are struggling to define, the first selves. I thank E. N. Trifonov for bringing up the important and often overlooked definition in reference 3.
References 1. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). 2. J. Joyce. in D. W. Deamer and G. R. Fleischaker (Eds.), the foreword of Origins of Life: The Central Concepts, Jones and Bartlett, Boston (1994). 3. A. I. Oparin. as referred to in: R. Popa. In Between Necessity and Probability: Searching for the Definition and Origin of Life. Series: Adv Astrobiol Biogeophys, Springer, NY, pp. 197-205 (2004). 4. E. N. Trifonov and V. Brendel. GnomicA Dictionary of Genetic Codes (1986), VCH. Balaban Publ., Weinheim, Germany. 5. C. Darwin. The life and letters of Charles Darwin (1888); Vol. 3, p. 18. Letter to Joseph Hooker. John Murray, London. 6. E. N. Trifonov. J Cosmology 10, 3374-3380 (2010). 7. S. Pino, E. N. Trifonov, and E. Di Mauro. Genomics, Proteomics and Bioinformatics 9, 7-14 (2011).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

Defining Life: An Exercise in Semantics or A Route to Biological Insights?

http://www.jbsdonline.com Asking What is X? questions is a natural human inclination, and because all scientists that have so far published are undoubtedly human, they discuss such issues often enough. What is Life? almost by definition is the king of such questions as far as Biology is concerned (1). Indeed, in the article which I address in this commentary (2), Edward Trifonov cites a recent special issue of the journal Origins of Life and Evolution of Biospheres that consists of 16 articles fully dedicated to different aspects of defining Life and the features of the resulting definitions (3). Certainly, this is evidence of considerable attention to the subject. Yet, in itself the question What is Life? hardly can be considered scientific. Falsification is impossible: whenever one finds an apparent counterexample, i.e. an entity that possesses all attributes included in the given definition that, however, is clearly not alive or conversely, an entity that lacks some of those attributes but is obviously a life form, some kind of intuitive understanding of the living state superseding any definition is involved. Even corroboration of a definition of life that would involve finding more and more diverse entities fitting the definition is compromised by the same problem: to count a case as supportive, an independent criterion is required, but this can only be intuitive. So we seem to know it when we see it but defining life is an elusive goal and apparently an inherently meta-scientific (metaphysical) task. The metaphysical character of the quest for the best definition of life does not necessarily imply that the exercise is futile. On the contrary, it is easy to envisage at least two areas of utility for such definitions: didactic better teaching of the fundamentals of biology and heuristic formulation of new falsifiable hypotheses and perhaps identification and study of novel life forms. I will not address the didactic aspect here but will briefly discuss the potential heuristic power of life definitions after first commenting on Trifonovs approach and conclusions. Trifonov goes about the derivation of a consensus definition of life in a manner that is unusual in scientific treatises but that has served him fairly well in previous work on the order of appearance of codons in the genetic code (4). The approach consists in compiling as many (supposedly) independent definitions as possible and then comparing vocabularies of these definitions to derive a consensus, the essential core that in itself may be hoped to provide for a better (the best) definition. There is no genuine scientific justification behind this approach and no guarantee that the numerous compared definitions are not all based on common misconceptions. In part, this indeed could be the case. Trifonovs core/consensus, additionally reduced through elucidation of apparent dependencies between some of the core terms, provides for a sensible, intuitively plausible minimalist definition: almost exact self-reproduction or self-reproduction with variations (2). It is certainly interesting, as Trifonov points

Eugene V. Koonin1
1

National Center for

Biotechnology Information, National Library of Medicine, National Institute of Health, Bethesda, MD 20894

Corresponding author: Eugene V. Koonin E-mail: koonin@ncbi.nlm.nih.gov

603

604
out, that this objectively derived minimalist definition of life almost exactly matches the definition given about 50 years by none other than Alexander Ivanovich Oparin, the famous even if notorious Russian biochemist who propounded the first concrete physico-chemical scenario for the origin of life. Yet, all its simplicity and appeal notwithstanding, the minimalist definition appears to be neither necessary nor sufficient, not even internally consistent. A simple implication of information theory (and more fundamentally, thermodynamics) is that error-free replication (more precisely, any information transmission process) is impossible (5). Hence the phrase self-reproduction with variation is actually redundant because any replication process will be characterized by some intrinsic error rate. The problem is exactly the opposite: it has been shown by Eigen and others that for stable information transfer (inheritance) down the chain of generations to be sustained, the error rate must not exceed a certain critical value known as error catastrophe or mutational meltdown threshold (6, 7). Thus, a necessary condition for life to evolve is not simply replication and not replication with variation (a tautology) but replication with an error rate below the sustainability threshold (Trifonovs almost exact selfreproduction fits the bill but appears imprecise). Another feature of a self-reproducing system that appears necessary for evolution is the phenotype-genotype feedback, or more precisely, differential effect of errors on the replication rate. Such differential fitness effect of mutations is a necessary condition of selection, both in its purifying and positive (Darwinian) incarnations. Hence replication with an error rate below the sustainability threshold, with non-uniformly distributed fitness effects of errors could be a candidate for a necessary and sufficient definition of life, or probably more precisely, a criterion for identification of life forms (5). Both Trifonovs consensus definition and the amended one given here are strictly informational. Any biologist will immediately feel that something is missing in these definitions. In broad outline, these missing components are: i) chemistry (metabolism), ii) energy conversion, and iii) structure (various forms of compartmentalization); for brevity, these may be denoted operational components of life forms, in contrast to informational components (8). Even more damning for the informational definitions of life, it may appear that these definitions are easily falsified by computer viruses and all forms of artificial life that replicate, mutate and evolve (9) but arguably are not actual life forms. However, such falsification is illusory for the simple reason that these evolving entities themselves are produced by highly evolved life forms. Clearly, the emergence of such life forms (and human civilization in particular) was a pre-requisite for the advent of these replicators they are not life forms as such but clearly are derived from life forms. We are currently unaware of life forms evolved from inanimate matter (or so we

Koonin
believe) that would lack any of the major operational attributes, in addition to replication. The operational components seem to be pre-requisite for the evolution of replicators fitting the above criteria, hence the informational definition of life presupposes the existence of some forms of metabolism, energy transformation and compartmentalization. However, all life forms on earth, with the exception of viruses which are obligate intracellular parasites, encompass a much stronger connection between replication, chemistry, energy and structure: the replicating moiety (genome) encodes key information on the operational components (primarily in the form of proteins). Is encoding operational components in the genome a necessary attribute of life? This does not appear certain at all, and here we come to the potential heuristic value of life definitions. Let us formulate a hypothesis: there are purely informational life forms in which genomes carry only the minimal information required for replication whereas all operational components are supplied by the (conducive) environment. Actually, hypothetical information-only replicators are among the essential features of several origin of life scenarios because emergence of complex genomes encoding operational components prior to the advent of an efficient replication mechanism appears effectively impossible (5, 10). This hypothesis may be hard to falsify but it certainly can be corroborated under two types of settings: in the laboratory and in extraterrestrial biospheres if and when such are discovered. Exploration of putative life outside earth belongs in the future but vigorous attempts to create in the laboratory an evolving system of replicators employing exogenous supplies chemicals and energy are underway, and certain progress has been achieved with ribozyme polymerases (11). However, these experimental systems remain a far cry from the efficient replicators required to start the evolution of life. Thus, the jury is still out on the information only life hypothesis. A remarkable feature of all known biological replicators is their digital character: these replicators are polymers consisting of multiple types of monomers. Such polymers appear uniquely suited for information encoding, so a plausible hypothesis seems to be that digital properties are necessary for life. This hypothesis certainly would be falsified by the (remarkable) discovery of analog life forms. To summarize, I believe that the democratic approach applied by Trifonov to the definition of life problem did not lead him far astray and converged on a natural and sensible, if not quite precise, informational definition. In my view, although life definitions are metaphysical rather than strictly scientific propositions, they are far from being pointless and have potential to yield genuine biological insights.
References 1. E. Schroedinger. What Is Life?: with Mind and Matter and Autobiographical Sketches Cambridge University Press, Cambridge (1992). 2. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Defining Life: An Exercise in Semantics or A Route to Biological Insights?

3. J. Gayon, C. Malaterre, M. Morange, F. Paulin-Cerceau, and S. Tirard. Origins Life Evol Biospheres 40, 119-244 (2010). 4. E. N. Trifonov. J Biomol Struct Dyn 22, 1-11 (2004). 5. E. V. Koonin. The Logic of Chance: The Nature and Origin of Biological Evolution FT press, Upper Saddle River, NJ (2011). 6. M. Eigen. Naturwissenschaften 58, 465-523 (1971).

605

7. J. Summers and S. Litwin. J Virol 80, 20-26 (2006). 8. R. Jain, M. C. Rivera, and J. A. Lake. Proc Natl Acad Sci U S A 96, 3801-3806 (1999). 9. C. Adami. Nat Rev Genet 7, 109-118 (2006). 10. E. V. Koonin. Ann N Y Acad Sci 1178, 47-64 (2009). 11. T. A. Lincoln and G. F. Joyce. Science 323, 1229-1232 (2009).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

Merits and Caveats of Using A Vocabulary Approach to Define Life

http://www.jbsdonline.com The field of defining life is rich in artful wording yet lacking in cohesiveness. Opinions about the necessity and possibility to produce a definition of life range widely. At the extremes (and without necessarily being wrong) some authors may claim that a technically accurate definition of life is not needed or impossible. Assuming that a definition for life is needed and possible, it has to obey two basic requirements: to be coherent relative to what we already know about life, and the philosophy used to produce the definition has to be clear of systemic errors. The method proposed by Edward Trifonov to define life (1) is a minimalist vocabularybased screening, combined with a personal interpretation of the meaning of the findings. Is this method consistent with the requirements listed above, and what novelties it introduces in our understanding of life? The method proposed is seemingly simple. Take a large collection of definitions of life and calculate the frequency of different words (Table I in the original article). Identify the most common words and the words with similar meaning (Table II in the original article) and combine them in a first-hand definition. Then, shorten by eliminating terms and concepts that imply each other, in a way that allows essence and causes rather than trivia and consequences to be retained in a final definition. Following this logic, the manuscript (1) should have ended with the analogy with Principal Component Analysis (p. 262). The author however finds more thrust to continue the manuscript past this point, by discussing issues such as thermostability, GC rich sequences, information complementarity and RNA-related almost precise replication. If the aim of this study (1) was to summarize why RNA-related molecules are important for biological life on Earth, then I find it insightful. If one however expected for this study to be a non-earth-centric attempt to define life, the focusing of the closing arguments on physical-chemical and informational properties of a particular class of molecules is distracting and unfulfilling. It is also necessary to analyze the validity of the key premise of this paper. It is obvious that the scientific community cannot bring itself together to produce and support a singular definition for life. The motives and the diversity of various opinions are not discussed here, only their consequences on a vocabulary-based strategy to define life. The essence of a vocabulary method is that words and ideas that are the most common must also be the most important. This is true to a point. It does apply very well to fields where basic research has more or less ended, yet it makes it difficult for pioneers and novel theories to gain recognition, irrespective of how right they are. For example, if we promote a scientific model based

Radu Popa
Portland State University, P.O. Box 751, Portland, OR, 97207, USA

Corresponding author: Radu Popa E-mail: rpopa@pdx.edu

607

608
on popularity, then Alfred Wegener was correctly ignored in 1912, when he promoted the concept of continental drift. One example from the field of the origin of life is Kunins contribution, which was used to explain the initiation of a living system as a symbiosis between two molecular networks (e.g. a protein-made RNA polymerase cooperating with an RNA-made protein polymerase) (2). This great contribution to understanding the essence of life would be considered unworthy of mentioning by a vocabulary-based method, simply because it is seldom cited in origin of life models. If we leave it to a computer-selected one-word-at-a-time vocabulary to define the essence of life then some seldom, yet worthy, characteristics of life would be dropped as little relevant. Take for example complex concepts such as adaptive evolution, cryptic information and energy dissipative systems. The combination adaptive evolution is richer in meaning than the term evolution. Evolution and variability exist in both alive and non-alive systems, yet living systems have an edge for survival by being built for adaptive evolution. Also, in all life forms that we recognize on earth today the genetic information is encoded. This is a very important attribute of life because it allows living entities to accumulate a collection of virtual realities (dissimilar states) that are all possible but do not interfere with each other, because when one is expressed all others are hidden. Obviously, the simple term information is too poor in meaning to describe the complexity of what life does. Lastly, if we discard the connection between the energy dissipative properties of living systems and their capacity for self-control we loose the very reason for the origin of life (3, 4). Can life be reduced to a collection of dimensionless qualities, or it has some properties that require reaching specific size before a system can become alive? The obvious example is the complexity-level analysis made by Stuard Kauffman (5), which a vocabulary analysis will simply ignore. The lesson learned from Dr. Trifonovs approach is that we can circumvent such limitations by analyzing major consequences of being alive rather than analytically dissect and list all lifes properties and achievements.

Popa
At the syntactic level a definition cannot be constructed as a popular saying that leaves grammatical parts in limbo. In order to be useful, the construction of a definition has to obey a specific set of minimal rules. The expression, Life is selfreproduction with variations albeit inspirational, does not inform whether life is a physical system or the property of a physical system. This definition does not clarify whether these features are restricted to life or if they may also occur in other systems that are not alive. Lastly, is this definition sufficient to explain life or is it just one important aspect of it? The analysis from (1) correctly replaces this truncated definition with a grammatically correct one developed by Oparin (6): Any system capable of replication and mutation is alive. This is not however sufficient reason to accept Oparins definition of life as complete, because it still lacks many important properties of life. Last but not least, can life exist that is not RNA-based? If the answer is YES, then no need exists to pound on the RNA-world drum in order to explain what life is. If NOT, then the author (1) has to state that no alien life can exist unless it is based on RNA-like molecules, which everybody will probably doubt. To summarize, one recognizes in a vocabulary method the need to resolve present ambiguities about defining life though, in my opinion, not the best avenue for reaching this goal. This aside, the reader interested in the subject of defining life and explaining its early evolution will find sufficient substance in (1) to make this article worth reading and instructive. Ultimately, We may never agree on a definition of life, which will remain forever subject to a personal perspective. The measure of ones scientific maturity may actually be his/her latest definition of life and the acceptance that it cannot be ultimate (7).
References 1. 2. 3. 4. 5. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). V. Kunin. Orig Live Evol Biosph 30, 459-466 (2000). P. Bak. How nature works. Copernicus, New York (1996). E. Chaisson. Int J Astrobiol 2, 91-101 (2002). S. A. Kauffman. The origins of order, Oxford University Press, New York (1993). 6. A. I. Oparin. Life: its nature, origin and development, Academic Press, New York (1961). 7. R. Popa. Orig Live Evol Biosph 40, 183-190 (2010).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

Self-Generated and Reproducible Dynamics in Gene Years Represent Life

http://www.jbsdonline.com In a recent article (1), Trifonov has developed an approach, conceptually inspired by principal component analysis (PCA), for arriving at a consensus definition of life as self-reproduction with variations. In appreciation of the sound analytical effort applied rigorously on a comprehensive body of literature, I will refer to both the approach and the definition as Trifonovian. The Trifonovian approach has to be commended in its effort to evolve a minimalistic, and at the same time an all-inclusive, definition of life. In fact, from the literature on molecular components of living systems, the Trifonovian definition of life is clearly a direct reflection of the ideas of consensus sequences at the level of DNA/RNA and (in specific cases) even proteins, and homology modeling (including minimal functional motifs) in protein structures. Thus, while appreciating the Trifonovian approach, it is also important to carefully consider the severe limitations of consensus/similarity approaches (applied for the molecular components) that have pushed the limits of experimental (both computational and wet-laboratory) biology into increasingly complex/sophisticated formalisms that have unfortunately not provided universal insights till date. In fact, some universal insights have been achieved only by comprehensive (computational and/or wet-laboratory) rather than consensus based approaches (2-5). Limitations of the Trifonovian Definition of Life A definition is expected to explain (and not just summarize) the meaning of a term. Scientifically, a definition has to be able allow extraction of parameters that will enable a mechanistic understanding of the term in view of the scientific method of Observation Hypothesis Experiment Mechanism. Thus, the obvious question arises why is a definition of life required? My answer to this question involves the following aspects A definition should allow (a) extraction of parameters, open to theoretical analyses and/or experimentation (computational/wet-laboratory), useful in providing mechanistic insights into life eventually leading to rules that can allow classification of a system as living or non-living, and, (b) a clear establishment of both necessary and sufficient requirements to be able to not just provide an understanding on the origin(s) of life but also lead to methodologies towards synthesis of life de novo. In this regard, the Trifonovian definition fails to address the above aspects either partially or in totality. For example, variation in the definition of life is subject to variation in the environment. If the environment was a static/stagnant variable, then variation in the definition of life would not be required (since variation in life is not observed in a static/stagnant environment). Further, the Trifonovian definition fails to address the molecular dilemma in realizing self reproduction within the constraints of conservation of mass and energy (i.e., exchanges with

Aditya Mittal
Kusuma School of Biological Sciences, Indian Institute of Technology Delhi, New Delhi 110016, India

Corresponding author: Aditya Mittal Phone: 191-11-26591052 E-mail: amittal@bioschool.iitd.ac.in

609

610
the environment must be built-in). On a trivial (and somewhat philosophical) level, the Trifonovian definition fails to classify sterile living systems (e.g., male mules) as a part of Life. Molecular Outlook for Developing a Definition of Life Extending the spirit of appreciating a requirement for the definition of life, let us carefully consider our understanding of a living system from the molecular perspective in terms of a unit of life, i.e. a cell. Figure 1A shows the basic molecular machinery of a unit of life. The molecular entities involved in giving a cell its identity as a living system in terms of

Mittal
the central dogma (6) are shown, though a few recently discovered exceptions to the central dogma have be reported recently (7-9). A living cell has evolved through presence of a single or several proteins resulting in phenotypic traits that are retained or lost in response to environmental changes. Interaction between the components of the proteome and the environment results in governing the dynamics and balance of the proteome to keep the cell as a unit of life in a given environment. Coding to construct this proteome is well guarded in two layers of code, neither of which readily interacts with the environment. Variations in this unit of life, in response to environmental changes, can result from either the feedback from the proteome to the genome, or, interaction between the

(A)

Environment

Phenotype Protein (Proteome)

Coded Message RNA

Code DNA (Genome)

Cell Unit of Life

Environment

(B)

Environment

Cell Environment 1

Phenotype Protein (Proteome)

Code/Coded Message RNA (Genome)

Virus Non-Living by itself Cell Environment 1

Environment

Figure 1: Molecular understanding of life and evolution. (A) A cell, considered (and defined) as a unit of life, is shown as a grey box. The molecular entities involved in giving a cell its identity as a living system in terms of the central dogma are shown. The environment is shown as a blue box. Components of the proteome interact with the environment (shown by bi-directional black arrow). Coding to construct this proteome is well guarded in two layers of code, neither of which readily interacts with the environment. Variations in this unit of life, in response to environmental changes, can result from either the feedback from the proteome to the code i.e., the Genome (shown by the uni-directional dashed black arrow), or, the more direct interaction between the code and the environment (shown by the bi-directional red dashed arrow). (B) An example of RNA virus (white box) is shown, where the code/coded message interact with two levels of environments (cell grey box: Environment 1, and the Environment of the cell/free-virus itself), in addition to the interactions of the proteome.

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Self-Generated and Reproducible Dynamics in Gene Years Represent Life

code and the environment. The former is responsible for accumulating variations over time leading to the slow process of evolution. For example, a protein not involved in interaction with the environment is gradually not present in a form to provide any feedback and thus is not further synthesized from the code. The resulting loss of this protein inside the cell gives rise to different phenotypic traits (variation). However, if the latter was possible, substantial variations would be observed in living systems at very short time scales. Even more drastically, in many cases life would not self-reproduce, since substantial variations would occur even with minor changes in the environment. Simply put, if the code was directly interacting with the environment, minor changes in the environment would yield sufficient changes in the code that could be amplified to form new life forms, rather than self-reproducing life forms. Every reproduction cycle would give rise to a new life form. Figure 1B shows an example of RNA virus, where the code/ coded message interact with two levels of environments, to support the above. For the virus, cell is the first level of environment that it has to interact with, referred to as Environment 1. However, since the virus has direct interaction of the code with its environment in addition to the proteome interacting with the environment, changes in the living form of virus is observed at very short time scales. This is true for all known RNA viruses (without exception). Thus, it is essential to appreciate that it is the phenotype that interacts with the environment, and only those living systems have survived (or will survive) in terms of self-reproduction in which the genotype (i.e., the code for creating phenotype) is not (or minimally) capable of interacting with the environment. Self-Generated and Reproducible Dynamics in Gene Years Represent Life Based on the above discussions, and inspired by the Trifonovian approach, it is desirable to utilize the comprehensive body of literature for arriving at a definition of life. However, it may be more useful to extract some universal concepts/principles discussed in Trifonov (1). It is clear that two key features must be built into a definition of life: Kinetics and Self-assembly.

611

Note that I have specified the two features in form of parametric variables that can be scientifically explored. Definition of life must specify its ability to self-generate (rather than selfreproduce with variations in response to the environment). The term self-generation accounts for a variety of self-assembly processes within living systems. The terms reproducible dynamics account for organized kinetic processes required for maintenance and self-generation properties of living systems. They also signify the importance of reproducing the dynamics, rather than the individual components behind the dynamics. In fact, at a molecular level, several features observed in and/or for living systems show dynamic behavior that, while stochastic in the actual initiation, follow clear and systematic kinetic modes from starting of an event to their completion regardless of the system and the technique (10-12). Finally, I would like to propose that analogous to the astronomical unit of light years (for length-scales in the universe) it will be useful to develop an evolutionary unit of gene years that can allow an understanding of time-scales for evolution of life and lengthscales for genomes in the proposed definition of life.
References 1. E. N. Trifonov. J Biomol Struct Dyn 29, 259-264 (2011). 2. G. N. Ramachandran, C. Ramakrishnan, and V. Sasisekharan. J Mol Biol 7, 95-99 (1963). 3. A. Mittal, B. Jayaram, S. R. Shenoy, and T. S. Bawa. J Biomol Struct Dyn 28, 133-142 (2010). 4. A. Mittal and B. Jayaram. J Biomol Struct Dyn 28, 443-454 (2011). 5. A. Mittal and B. Jayaram. J Biomol Struct Dyn 28, 669-674 (2011). 6. F. Crick. Nature 227, 561-563 (1970). 7. C. Kimchi-Sarfaty, J. M. Oh, I.-W. Kim, Z. E. Sauna, A. M. Calcagno, S. V. Ambudkar, and M. M. Gottesman. Science 315, 525-528 (2007). 8. M. Sharma, V. Hasija, M. Naresh, and A. Mittal. J Biomed Nanotechnol 4, 44-51 (2008). 9. F. Zhang, S. Saha, S. A. Shabalina, and A. Kashina. Science 329, 1534-1537 (2010). 10. A. Mittal, E. Leikina, J. Bentz, and L. V. Chernomordik. Anal Biochem 303, 145-152 (2002). 11. A. Mittal, E. Leikina, L. V. Chernomordik, and J. Bentz. Biophys J 85, 1713-1724 (2003). 12. T. Gattegno, A. Mittal, C. Valansi, K. C. Q. Nguyen, D. H. Hall, L. V. Chernomordik, and B. Podbilewicz. Mol Biol Cell 18, 11531166 (2007).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

A Minimal or Concise Set of Definition of Life is Not Useful

http://www.jbsdonline.com Trifonovs interesting analysis (1) appeared to have highlighted a minimal set of popular vocabulary in the literature that constitutes a concise and inclusive definition for life. Defined as such, Life is self-reproduction with variations. The linguistic and philosophical worth of the paper notwithstanding, its scientific value is, however, questionable. This minimized definition of life misses out on an important property, namely lifes capacity for integrating chemical processes that sustains the living entitys low entropy (i.e., metabolism). Importantly, the definition has no bearings on the origin of life (2-6), and ignored the myriad of possible transitions from a collection of abiotic chemical reactions to a form that could be subjected to Darwinian selection (7, 8). Though undoubtedly concise enough, it is far from being inclusive. A quick glance at the definition of Life is self-reproduction with variations brings to mind two potential deviations, both of which might nonetheless be recognized as life. Firstly, obligate parasites (the simplest of which being viruses) are in the strictest sense incapable of self -reproduction. Intriguingly, the number of viruses on the planet exceeds that of cells almost by an order of magnitude (9), and the pivotal role of viruses in the evolution of the biosphere is now increasingly recognized (10). Secondly, terminally differentiated cells, of which neurons would be prime examples, are incapable of replication. Neurons are nonetheless metabolically active, long living entities that last the life span of the animal. For that matter, senescing eukaryotic cells, which had exited the cell cycle and could no longer undergo cell division, could remain metabolically active for a good length of time. On the other hand, at least from a biocentric perspective, a selfreplicating computer virus or a Von Neumann probe (11) programmed to allow variations or mutations in their replications, which are not usually considered as leading candidates for life forms, could have their candidature legitimized by Trifonovs minimized definition. Although the author deemed the generality of the minimized definition as being an unforeseen and favorable property, it could just as easily be viewed as an undesirable oversimplification. A more disappointing feature of the minimized definition is its failure to connect with the problem of the origin, or emergence of life. Although probably not by design, results of the authors linguistic analysis appear to bias towards the gene-first or replicator-first school of thought, and have rather marginalized the metabolism-first theories. Various forms of the gene/replicator-first theories, including the RNA world (12), depict the emergence of a self-replicating molecule and a self-replicating process that could be subjected to Darwinian selection. The metabolism first models, on the other hand, emphasize on the

Bor Luen Tang

Department of Biochemistry, Yong Loo Lin School of Medicine, National University of Singapore, MD7, 8 Medical Drive Singapore 117597, Republic of Singapore

Corresponding author: Bor Luen Tang Phone: 65 516-1040 Fax: 65 6779-1453 E-mail: bchtbl@nus.edu.sg

613

614
importance of self-sustaining prebiotic chemical reactions that culminates into the earliest form of metabolism. Examples of these include the classical Wchtershusers Iron-sulfur world (13), and later proposals such as the thermodynamically sustainable autocatalytic cycles as expounded by Shapiro (14). Both schools of thought have merits in certain regards, while harboring their own shortcomings. These have been extensively discussed elsewhere (8, 15, 16) and shall not be elaborated. Suffice to note here that the minimized definition fails to capture the large variety and possibility in terms of prebiotic pro-metabolic reactions that are chemically conceivable, and their transitions into Darwinianly selectable life. A further consideration of this disconnection between this minimized definition with lifes origins concerns the sites involved. In a sense, the exact physical and chemical properties of life could be situation and locale dependent. On this planet itself, life could have emerged more than once (18), and at different geological sites (7, 8). The minimized definition fails to illustrate the myriad of possibilities of lifes emergence in space and time such as those from the hot hydrothermal vents of early Earth 4 billion years ago to the warming organic lakes of Titan another 4 billion years from now (17). Along a similar line of thought, I could not share the authors optimism that day is not far away when life could be created in a test tube. Never mind that Spiegelmans monster (19) represents a state of degenerated complexity that is unlikely to eventually acquire a degree of respectable, self-sustaining metabolism that is capable of adaptation and evolution, all other attempts to model life to date have involved starting materials that are far too complex that would have been available prebiotically. There is no evidence that a MillerUrey type experiment ever yielding polymers of a significant length, not with the opportunity and rate of a chain terminating reaction far exceeding that of spontaneous polymerization. In Kompanichenkos terms, we have no idea of the condition(s) that would generate oscillating prebiotic microsystem at a balanced bifurcate state, let alone those that would facilitate

Tang
inversion of the balance free energy/entropy contribution and the rise of the key biological properties (20). In view of the above, it seems reasonably clear that a popular or consensus minimalistic definition of life may not be very useful, and could in fact be misleading. This notion does not undermine the two principal features of the definition, namely self-reproduction and variations (which implies a propensity to undergo Darwinian selection), but rather suggests that our understanding of life in the universe might actually benefit from a broad definition with a wide range of vocabularies. A more wordy definition may appear more cumbersome, but its worth the trouble. In the context of Trifonovs paper, definition (1) is more useful than definition (2). The latter, like Spiegelmans monster, is less exciting.
References 1. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). 2. J. W. Szostak. J Biomol Struct Dyn 28, 1059-1059 (2011). 3. U. J. Meierhenrich, J.-J. Filippi. C. Meinert, J. H. Bredehft, J.-i. Takahashi, L. Nahon, N. C. Jones, and S. V. Hoffmann. J Biomol Struct Dyn 28, 1060-1061 (2011). 4. G. Costanzo, S. Pino, F. Ciciriello, and E. Di Mauro. J Biomol Struct Dyn 28, 1061-1061 (2011). 5. J. P. Ferris, P. C. Joshi, M. F. Aldersley, and J. W. Delano. J Biomol Struct Dyn 28, 1062-1062 (2011). 6. A. E. Engelhart, R. Buckley, E. D. Horowitz, and N. V. Hud. J Biomol Struct Dyn 28, 1063-1063 (2011). 7. B. L. Tang. J Br Interplanet Soc 58, 218-222 (2005). 8. B. L. Tang. Prog Nat Sci 17, 500-510 (2007). 9. A. Abroi and J. Gough. BioEssays 33, 626-635 (2011). 10. L. P. Villarreal and G Witzany. J Theor Biol 262, 698-710 (2010). 11. R. A. Freitas, Jr. J Br Interplanet Soc 33, 251-264 (1980). 12. W. Gibert. Nature 319, 618 (1986). 13. G. Wchtershuser. Prog Biophys Mol Biol 58, 85-201 (1992). 14. R. Shapiro. Q Rev Biol 81, 105-125 (2006). 15. R. Shapiro. IUBMB Life 49, 173-176 (2000). 16. J. Perat. Int Microbiol 8, 23-31 (2005). 17. R. D. Lorentz, J. I. Lunine, and C. P. McKay. Geophys Res Lett 24, 2905-2908 (1997). 18. P. C. Davies and C. H. Lineweaver. Astrobiology 5, 164-163 (2005). 19. D. R. Mills, F. R. Kramer, C. Dobkin, T. Nishihara, and S. Spiegelman. Proc Natl Acad Sci USA 72, 4252-4256 (1975). 20. V. N. Kompanichenko. 4 Int J Astrobiol 7, 27-46 (2008).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

On the Misgivings of Anthropomorphic Consensus Polling in Defining the Complexity of Life

http://www.jbsdonline.com In a recent paper (1) Edward N. Trifonov has turned his statistical expertise on one of the most enigmatic questions still vexing our scientific conceptions of the universe: What is life? If applied ingeniously, statistical methods are marvelous tools indeed, but yet, they hardly breed miracles on their own. To be sure, professor Trifonov has done wonderful work before, scrutinizing the system of coded protein synthesis for innate secrets by sophisticated statistical analyses (2-4). To start with, many others have put forth various definitions of the living state as such in contrast to non-living matter, which pervades most of the observable universe. By tabulating 123 versions of such learned statements (partly overlapping, that is, but slightly or distinctly different from one another), and to simplify matters, Trifonov (1) has now applied his statistical tools on the wording of these definitions. Would the grouping of major terms by related meaning, as followed by word counts of occurrence in the major groups, yield any superior, canonical relationship? Nota bene, not just a meaningful concept at a superior level, but a universally applicable, novel and minimalistic definition of life as such, to the inclusion of any forms of life imagination may offer? Although he does not meticulously discern what properties should be required of such a universal definition, Trifonov proposes to have found exactly that, as merely based on just two terms: Life is self-reproduction with variations. In fact, this punchline was put forth earlier (4), as an aphoristic rephrasing of Darwins principle. The author must have found it reassuring that an earlier hunch of his could be extracted as a consensus from a much larger sample from the expert literature, although only ~15% (18 of 123) happened to contain a pair of terms related to his condensate. Moreover, by explicitly referring to Oparin [as cited in (5)] Any system capable of replication and mutation is alive Trifonov moves close to implying that his minimalistic definition may not only be necessary but even sufficient for sorting out the living state from non-living matter. On the other hand, he likens his statistical vocabulary approach to a Principal Component Analysis (aimed to reduce a complex data set to an array of subsets that vary independently of one another) with the extraction of a single principal component (the major one, covering most of the data), which might serve as the main characteristic of the living state so far, so good. Yet, as the author himself is quite aware, Even most primitive forms of observable life are still too complex, to claim that they can be reduced to the above simple formula (1). Hence, it is rather premature to adopt his simple formula as the defining principle of any forms of life imagination may offer . As to my personal imagination, for that matter, I should not hesitate to subsume frost

Richard Egel
Department of Biology, University of Copenhagen Biocenter, Copenhagen, Denmark

Corresponding author: Richard Egel E-mail: regel@bio.ku.dk

615

616
tracery on a window pane or frostwork-type mineralizations in cave deposits under this generally defined set of natural phenomena. They actually replicate certain preexisting patterns, and do so with many variations, but calling them life-like for more than a rather superficial resemblance should be besides the point, I guess. There are good reasons to suggest, therefore, that Trifonov should not stop at the very first principal component of his statistical vocabulary filtering approach. Additional components, likewise both principal and indispensable for zooming in on the essence of life, should not lightly be dismissed beforehand. Alternatively, the intended meaning of the main part in his favorite component self-replication should be specified so distinctly and concisely that no form of nonliving matter imagination may offer might ever qualify for having that capacity. If, and only if this precondition has been met can self-replication be considered equivalent conceptually and empirically to life as such. And what a big if that is, since now the burden of defining the complexity of life in simple terms has shifted to defining self-replication and its preconditions at the same level of life-like complexity. In this brief discourse I have deliberately focused on the main part of the composite formula, Life is self-reproduction with variations, since the secondary part does not add a new dimensionality to the problem at all. To be precise on this, self-reproduction without variation would be entirely fictitious, in that it can never be realized as a natural process. In fact, in the course of biological evolution of earthly life as we know it the multiple sources of stochastic variation in reproductive processes have progressively been narrowed down to acceptable and affordable levels, but eliminating each of such sources once and for all is beyond the bounds of possibility. Why should the essence of life as such be defined in idealistic or anthropomorphic terms in the first place? Who would actually benefit from such a canonical definition if it ever existed? While most biologists in general do not care, it is foremost some non-biologists trying to comprehend the enigmatic origins of life from non-living matter who are convinced that their goal cannot be reached without having defined life as such beforehand. Yet, not all such scholars share this conviction, and personally I very much tend toward the sceptical side on this dividing issue (6). In defence of this sceptical stance, I consider it more relevant for a comprehension of biological complexity to familiarize with

Egel
the principles of fuzzy sets and fuzzy logics (7). This should not be denounced as fuzzy thinking in a pejorative sense. Rather, it is to appreciate the appropriateness of variance-loaded terms and categories for about every aspect of the natural world, including life with all its intrinsic complexity and evolutionary history. At the initial stages, lifes emergence had to cope with much higher degrees of variance than what is presently observed in the current biosphere. Hence, putting coordinate restraints on stochastic variance in ever so many independent categories by natural selective processes should be at the forefront of how to understand the living state. To illustrate the anthropomorphic character of certain unreflected definitions, I should like to draw on a comparison of German and English conventions concerning food intake. While it should sound sensible enough in German to define the human species as das essende Wesen, the direct translation into English would make no sense: the eating creature would cover most animals alike. British pragmatism, at least on this issue, allows animals at large to eat their food as well, whereas essen in German is a decidedly human privilege, in contrast to fressen, as used for the equivalent activity in animals. Such a pseudo-definition, however, would be rather trivial indeed, a human being is a human being, Period!. In summary, the statistical vocabulary approach of Tifonov (1) to extract a simple defining formula for the intrinsic complexity of life amounts to an enchanting exercise on the border between basic science and aphoristic poetry. I was somewhat reminded of my first visit to the United States in the mid sixties, when a frenzy florished among high school kids to come up with the most fanciful variation on Happiness is ....
References 1. 2. 3. 4. 5. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). E. N. Trifonov. Gene 261, 139-151 (2000). E. N. Trifonov. Phys Life Rev 5, 121-132 (2008). E. N. Trifonov. Res Microbiol 160, 481-486 (2009). R. Popa. In Between Necessity and Probability: Searching for the Definition and Origin of Life. Series: Adv Astrobiol Biogeophys, Springer, NY, pp. 197-205 (2004). 6. R. Egel. Integrative perspectives: In quest of a coherent framework for origins of life on earth. In Origins of Life: The Primal SelfOrganization. Egel, R., Lankenau, D.-H., and Mulkidjanian, A. Y. (Eds.), Springer-Verlag, Heidelberg, GE, pp. 289-360 (2011). 7. G. Bruylants, K. Bartik, and J. Reisse. C R Chimie 14, 388-391 (2011).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

The Complexity of Life: Can Life be Simply Defined?

http://www.jbsdonline.com In the interesting article by Trifonov (1), the definitions of Life were analyzed from a variety of sources and 10 groups of terms were compiled. Based on these terms, the author attempts to provide a succinct definition of what Life is; that is, Life is self-reproduction with variations. While the terms underpinning the definition clearly provide a reasonable characterization of Life, there appears to be a lack of certain key elements of what Life embodies. One of the main issues with the definition as based on the vocabulary used is that there is no mention that Life, as it had evolved (2-6) on earth, came about without a purpose. One of the definitions the author had quoted was from Kompanichenko who stated that the category of terms that encapsulates the definitions of Life is one that includes, among other characteristics, the display of self-perfecting logic. First of all, there is a slight problem with the term self-perfecting, as it tends to give the impression that there is a teleological attempt by Life to drive towards a particular aim such as survival or an increase in complexity. Secondly, words under the category of Ability more often than not promote the perception that living things are consciously able to respond physiologically to external environment and to evolve accordingly, rather than there being a selective pressure exerted by the external environment, thereby sieving out organisms not suited for survival in that particular circumstance. As argued by Richard Dawkins (7), Life evolved without an aim or a purpose. As such no pre-determined conditions existed that favoured a particular life form over another or any life forms for that matter. Conversely, organisms are not capable of evolving specifically to adapt to various conditions on Earth. The process of evolution, as expressed by Franois Jacob (8), is one of tinkering, during which the selective process works on what already exists, either transforming a system to give it a new function or combining several systems to produce a more complex one. That vestigial organs or lack of a perfect design in organisms (9) can be found today supports this notion. While the author (1) was relying on terms previously used to derive a concise definition of Life based on the frequencies of their usage, I would suggest that there could be a description to reflect Life as an entity or property that is amenable to (natural) selection without a purpose (direction). In addition, words such as organization and information under the groups system and complexity do not completely embody the idea of emergent properties that relate to developmental biology at the individual organism during embryogenesis, as well as for the entire collection of living organisms. As put forth by Oyama (10), there appears, at least during the development of a multicellular organism, to be more than the information encoded by genes. Indeed, the interaction among cells is critical during development to ensure the proper unfolding of the resulting

Foong May Yeong

Department of Biochemistry, Yong Loo Lin School of Medicine, MD7, 8 Medical Drive, Singapore 117597

Corresponding author: Foong May Yeong E-mail: foong_may_yeong@nuhs.edu.sg

617

618
organism. Such an important essence of Life, where invariant properties such as information encoded in the genes and how cells interact with its environment determine how multicellular organisms develop, might need to be included in the definition of Life. Likewise, the terms under the section on evolution appear to fall short of representing the full complexity of how organisms can evolve as part of being a living thing. Currently, there is a trend towards examining the evolvability of organisms as an extension of the ideas behind the Modern Synthesis of evolutionary biology (11). This is based on the authors count of 364 papers in which this word had been used (11). Simplistically, evolvability can be loosely referred to as the propensity of a genotype to generate variability and as such, determine the success of a species during natural selection. While the notion of evolvability has not been formally defined and empirically validated, based on the authors argument, I suggest that the concept might eventually be a more useful one to explain mechanistically how different species of organisms could have arisen. This term does not seem to be encapsulated in the short definition of Life that is provided (1) and it is not clear if it could be in the future. Furthermore, the definition should encompass the notion that while the various properties of Life as complied (1) are carried within each living organism, it is the collective existence all living things as a result of organisms interacting with the environment that we are able to witness the phenomenon of Life.

Yeong
The definition of Life is important as it is needed for philosophical and practical reasons (12). Given the complex nature of Life, it might require a more elaborate definition (13) to cater to specific applications (14) than simply being distilled from the vocabulary of previous definitions of Life to be of practical use.
References 1. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). 2. W. L. Duax, R. Huether, D. Dziak, and C. McEachon. J Biomol Struct Dyn 28, 1066-1067 (2011). 3. A. Y. Panchin, E. N. Shustrova, and I. I. Artamonova. J Biomol Struct Dyn 28, 1068-1068 (2011). 4. L. D. Williams, C. Hsiao, J. C. Bowman, C. R. Bernier, J. Peters, D. M. Schneider, and E. ONeill. J Biomol Struct Dyn 28, 1071-1072 (2011). 5. T. J. Glembo and S. B. Ozkan. J Biomol Struct Dyn 28, 1068-1069 (2011). 6. A. Goncearenco and I. N. Berezovsky. J Biomol Struct Dyn 28, 10651065 (2011). 7. R. Dawkins. The Blind Watchmaker: Why the Evidence of Evolution Reveals a Universe without Design, W. W. Norton & Company (1996). 8. F. Jacob. Science 196, 1161-1166 (1977). 9. S. J. Gould. The Pandas Thumb: More reflections in Natural History, W. W. Norton & Company (1982). 10. S. Oyama. The Ontogeny of Information, Duke University Press (2000). 11. M. Pigliucci. Nat Rev Genet 9, 75-82 (2008). 12. J. Gayon. Orig Life Evol Bioshp 40, 231-244 (2010). 13. R. Popa. Orig Life Evol Bioshp 40, 183-190 (2010). 14. J. D. Oliver and R. S. Perry. Orig Life Evol Bioshp 36, 515-521 (2006).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

The Role of Logic and Insight in the Search for a Definition of Life
http://www.jbsdonline.com In a recent paper Trifonov (1) has carried out a statistical analysis of the frequency of the vocabulary used in 123 existing definitions. The goal is to identify the most important words and to use these for phrasing a commonly acceptable definition of life. Trifonov arrives at the conclusion that (self-)reproduction and evolution form the minimal set for a concise and inclusive definition: Life is selfreproduction with variations. Also a more lengthy definition is presented: Life is metabolizing material informational system with ability to self-reproduction with changes (evolution), which requires energy and suitable environment. Trifonovs work is part of the important and long standing quest in science and philosophy for a commonly acceptable definition of life (2-7). While reading Trifonovs publication a few questions came to my mind. The first question is why the author suggests using a vocabulary method instead of insight when defining life? I am worried that although the use of vocabularies may represent a proper tool for identifying keywords and the like, the methodology seems fundamentally inappropriate for suggesting definitions. Would any definition process not require the logical integration of scientific insights and thorough testing by confronting them with difficult cases? It is not clear to me how vocabulary studies meet such criteria, because the ranking of words according to frequencies seems blind to the underlying logical relationships. My second question involves some methodological aspects of using vocabulary analyses for creating definitions. 1. Can lists of definitions reflect recent developments? Innovative insights will in general require several years to become generally referenced in the literature. In addition, recent studies will generally relate to advanced scientific insights and may not use conventional wording. Would this effect potentially bias lists of definitions towards past and potentially aged insights? 2. Has the list of definitions been checked for dependence of information? It is general practice to use old masters as a basis when improving definitions. Such practices are likely to cause biases towards certain words in definitions inspired by the most influential examples. 3. How to extract meaningful results if vocabulary studies take words literally? Definitions frequently originate from different worldviews. As a consequence, the meaning of the words may differ between definitions. For example, some people use life for indicating all ecosystems

Gerard A. J. M. Jagers op Akkerhuis

Center for Ecosystem Studies, Alterra, Wageningen University, 6700 AA Wageningen, The Netherlands

Corresponding author: Gerard A. J. M. Jagers op Akkerhuis E-mail: gerard.jagers@wur.nl www.hypercycle.nl

619

620
and organisms on earth (life on earth), while others think of the length of the period between birth and death (he had a long life) or of daily existence (my life as a teenager). How has the study accounted for the summation of words with different meanings? 4. Would the vocabulary method recognize the confusion that presently exists with respect to the word life? Many well-known definitions of life actually refer to concepts related to living, for example metabolism, activity, reproduction, etc. As has been suggested by the author of this comment, it may be profitable to science when life would be used to describe organisation, while living is used in relation to the dynamics of those organisations representing life. This distinction was already recognized by the famous Socit de Biologie in Paris (8). Examining frozen or dried bacteria, the Society concluded that the potential to revive an anabiotic stage is an inherent aspect of the organisation of the material of which the object consists and that it is equally persistent as the molecular state of the matter forming the system. The society concluded: La vie, cest lorganisation en action. In other words, living refers to the dynamics of organization(s) that represent life. How would vocabulary studies assist in identifying which organisation(s) represent life? 5. How do vocabulary studies distinguish between appropriate and inappropriate definitions? Without such distinction, the most popular concepts will always be partly associated with inappropriate definitions. How can statistics on vocabulary circumvent the problem of deciding about the correctness of a given definition? My third question involves the testing of definitions based on vocabulary analysis. Any commonly acceptable definition must be able to deal with difficult cases. What happens if we undertake this exercise using the above definitions? How about a sterilized cat? If I understand the above definitions well, this cat would not be a living being because it cannot self-reproduce. Even a normal fertile cat is a problem. It cannot self-reproduce, because it needs the males sperm. And a frozen bacterium? While being frozen, it is neither metabolically active nor can it reproduce. Not life according to the above definitions. But a frozen bacterium still possesses the structure of life, which can be demonstrated when it resumes living activity after being thawed. How to value the above definitions in the light of these results? In relation to the above questions, I would like to invite the author to take an interest in a recently developed framework that uses the evolution of complexity as a basis for defining life. In order to analyse the organisation in nature

Jagers op Akkerhuis
in a stringent way, the author of this comment developed the Operator Hierarchy (9, 10). This hierarchy represents a ladder ranking all types of physical particles and types of organisms, generically indicated as operators, according to discrete transitions in the complexity of their organisation. As has been advocated in a previous publication about the definition of life (11), this ladder offers a fundamental basis for defining life as a common property of all types of entities on the ladder (the operators) that are equally or more complex than the cellular operator (bacteria s.l.). From this point of view, talking about life implies a focus on the presence of the level-defining organisation in selected operators. As long as the level-defining organisation is present, the entity represents life. And when an entity that represents life is dynamically active, it is living. Consequently, death implies the loss of the level-defining organisation. Biologists generally consider all the operators with a minimum complexity of the (bacterial/prokaryotic) cell as organisms. The complexity ladder of the operator hierarchy thus offers an underlying logic connecting the concepts of life and the organism. The ladder therewith solves an old circularity problem that occurs when life is referred to as a property of organisms, while organisms are defined as living beings. As has been discussed in (11), the operator based definition of life deals without problems with a broad range of difficult cases. Assuming that defining life requires the logical integration of scientific insights and thorough testing of the results by confronting them with difficult cases I was inspired to a range of questions about specific aspects of the use of vocabulary study for defining life. I find it worrisome that the method seems not to invoke insight and that the resulting definitions seem to have problems with simple test cases. I hope the author can take away my worries in his response.
References 1. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). 2. R. Popa. Between necessity and probability. Searching for the definition and the origin of life. Advances in Astrobiology and Biogeophysics. Springer (2003). 3. D. E. Koshland, Jr. Science 295, 2215-2216 (2002). 4. C. Emmeche. Ultimate Reality and Meaning 20, 244-264 (1997). 5. M. A. Bedau. What is life. In: Sarkar, S. and Plutynski, A. (Eds.), A companion to the philosophy of biology, pp. 455-603 (2007). 6. C. E. Cleland and C. F. Chyba. Does life have a definition? In: Planets and Life: The Emerging Science of Astrobiology, Woodruff, T., Sullivan, I. I. I., and Baross, J. A. (Eds.), Cambridge University Press (2007). 7. J. M. Maynard Smith and E. Szathmry. The origins of life. Oxford University Press, Oxford (1999). 8. P. Broca. Mm Soc Biol 3me Srie II, 1-139 (1860). 9. G. A. J. M. Jagers op Akkerhuis and N. M. van Straalen. World Futures, the Journal of General Evolution 53, 329-345 (1999). 10. G. A. J. M. Jagers op Akkerhuis. The operator hierarchy, a chain of closures linking matter, life and artificial intelligence. PhD thesis, Raboud University, The Netherlands (2010). 11. G. A. J. M. Jagers op Akkerhuis. Found Sci 15, 245-262 (2010).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

Life 5 Self-Reproduction with Variations?

http://www.jbsdonline.com Trifonovs paper (1) is a delightfully clever, objective and quantitative approach to defining life. Doing a linguistic analysis of 123 published definitions of life, Trifonov tabulates the words used in these definitions, to seek a consensus definition of life. This approach synthesizes some of the thinking and work of hundreds of scientists, and non-scientists as well. Popa (2) contributes 90 of the definitions used by Trifonov. Popas list of 90 definitions of life is an unusual one. Historically, it is impressive, running from 1855 to 2002. It is also very broad, including non-scientists such as Friedrich Engels [No physiology is held to be scientific if it does not consider death an essential factor of life. . . . Life means dying. From Dialectic of Nature (3)] and unusual definitions, such as: Life is a system which has subjectivity. Popa himself says that his list of lifes definitions serves only a general bibliographic purpose, and he cites four bibliographies and discussions of definitions of life that he says are more extensive than his own. Would any new insights be gained by doing a similar analysis of more rigorous bibliographies of the definitions of life, such as the four cited by Popa? Would it make sense to weight recent definitions of life more heavily than older definitions of life, given that we have learned much about life in recent decades? For example, in 1944, when Erwin Schroedinger wrote his book, What is Life? (4), he predicted that life will be found working in a matter that cannot be reduced to the ordinary laws of physics. Life is Self-Reproduction with Variations. This consensus definition of life from Trifonovs paper (1) is reduced to only two concepts Self-Reproduction and Variation. It contrasts with the often-used definition from a panel for NASA (National Aeronautics and Space Association): Life is a chemical system capable of Darwinian evolution. (5) In a paper based on this definition, Benner (6) explains how reproduction with variation is not an acceptable definition of life, because crystals grow, incorporating defects; and they reproduce when powdered and used to seed the growth of more crystals. What crystals lack is heritability, as in Darwinian evolution. Perhaps a better definition of life would be, Self-reproduction with heritable variations. In Popas 90 definitions, I find only about 17 instances that might be related to heredity, by summing the following search results: heritable, heredity, hereditary, genetic, anagenetic, genome, mutation, and Darwinian. Perhaps there are other relevant search terms that I have not thought about, or perhaps heritability is simply not a common concept for defining life. What about viruses? There is good reason to argue that viruses are living parasites that are capable of reproduction but not self-reproduction. This classification

Helen Greenwood Hansma

Department of Physics, University of California, Santa Barbara, CA 93106

Corresponding author: Helen Greenwood Hansma E-mail: hhansma@physics.ucsb.edu (or) helen.hansma@gmail.com

621

622
of Viruses as Living is consistent with Trifonovs linguistic analysis, in which there are 25 instances of reproduce/reproduction/replication but only 5 instances of self-reproduction. Whether or not viruses are alive, they are clearly on a continuum between Living and Non-Living. The existence of such a continuum complicates the search for an all-purpose definition of life. Do we need an all-purpose definition of life? On one hand, Dyson writes an entire book, Origins of Life (7), without explicitly defining life. Origins-of-life researchers do not necessarily need a definition of life. For example, when I write about the possible emergence of life between mica sheets (8), I am hypothesizing about a wide span of events from nonliving to living. On the other hand, Hazen (9) says scientists crave an unambiguous definition of life. In response to the problem of defining life, another 2011 paper (10) uses quite a different approach. This paper rejects altogether attempts to define life, saying that the concept of life is impossible to define and is thus a metaphysical concept instead of a scientific concept. The proposed solution is to use origin of evolution instead of origin of life. Evolution, it says, may be defined by as few as three conditions: [1] the emergence of open non-equilibrium structural systems, [2] self-replication, and [3] the acquisition of heritable structure/function properties. Self-replication is comparable to Trifonovs self-reproduction, though replication is perhaps a worse term, because it is typically used to describe the copying of genetic material, which is only one element in the reproduction of an organism. Origin of Evolution is especially problematic, given the use of evolution in other contexts, such as cultural evolution, evolution of language, and evolution of the airplane. One is left with the question, Origin of Evolution of What? In fact, a new scientific subfield is the evolution of minerals. (11) Life and minerals have co-evolved, according to Hazen, who states that the earth had ~1500 minerals before the origin of life, increasing to ~4300 minerals today, most of which may be the result of biochemical processes. The origin of evolution

Hansma
might thus be said to start when the biogeochemical processes started during the process of the origin of life. One of the biggest needs for a definition of life is in the field of exobiology. What does one look for, when seeking evidence for life on Mars, for example? We know about life on earth, but what about life as we do not know it? Trifonovs paper is clearly useful for determining whether life has been created in vitro, as he describes in his paper, that cross-replicating ribozymes (12) evolve but do not replicate. Does Trifonovs paper help those who are searching for life on Mars? Probably not. Benner (6) addresses this problem by going beyond the definition of life and into a range of questions, such as, Does life require carbon? And, does life require water? Regarding the first question, the response is that carbon forms stronger bonds than silicon, but only about one-third stronger. In summary, Trifonovs paper is not the final answer to the question of what is life. It is, however, a brilliant approach to the problem of giving scientists and non-scientists an unambiguous definition of life.
References 1. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). 2. R. Popa. Between Necessity and Probability: Searching for the Definition and Origin of Life. Adv Astrobiol Biogeophys (SpringerVerlag, Berlin, 2004), pp. 197-205. 3. R. Popa. Between Necessity and Probability: Searching for the Definition and Origin of Life. Adv Astrobiol Biogeophys (SpringerVerlag, Berlin, 2004), pp. 227-252. 4. E. Schroedinger. What is Life? & Mind and Matter (University Press, Cambridge, 1944). 5. G. F. Joyce. In Origins of Life: The Central Concepts, in D. W. Deamer and G. R. Fleischaker (Eds.), (Jones and Bartlett, Boston, 1994). 6. S. A. Benner, A. Ricardo, and M. A. Carrigan. Current Opinion in Chemical Biology 8, 672-689 (2004). 7. F. J. Dyson. Origins of life. (Cambridge University Press, Cambridge [England]; New York, ed. Rev., 1999). 8. H. G. Hansma. Journal of Theoretical Biology 266, 175-188 (2010). 9. R. M. Hazen. Genesis: The Scientific Quest for Lifes Origin. (Joseph Henry Press, Washington, DC, 2005). 10. M. Tessera. Int J Mol Sci 12, 3445-3458 (2011). 11. R. M. Hazen, et al., American Mineralogist 93, 1693-1720 (2008). 12. T. A. Lincoln and G. F. Joyce. Science 323, 1229-1232 (Feb. 27, 2009).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

Life: Self-Directing with Unlimited Variability on Self-Speeding

http://www.jbsdonline.com In his recent report in this journal, E. N. Trifonov (1) suggests that life should be defined as self-reproduction with variations, based on a vocabulary analysis of previous definitions of life. The method is novel, and the conclusion is interesting. However, there are some problems in the analysis process, and the final assertion of the definition of life needs more cautious and deeper consideration. The initial analysis, especially concerning the nine word groups (so-called definientia), is impressive, which in some degree could reflect peoples common views up to now on the essential meaning of life phenomenon. This analysis leads to a collective definition of life as metabolizing material informational system with ability of self-reproduction with changes (evolution), which requires energy and suitable environment (referred to as definition [1] in the original paper). As the author says, this collective definition, though comprehensive, should be worked on to reach a more concise one. This work is important because a satisfying definition should not be a collective one. However, the author does this work in a way of some curtness. For example, it is mentioned that metabolism implies both energy and material supply which also represent environment. Though this seems true, on the purpose of taking out the words energy, material and environment from the list, the author should say more than just one sentence, especially considering metabolism is a word with quite different interpretations. Furthermore, it seems that self-reproduction (replication) implies metabolism as well needs a more detailed explanation rather than the self-reproduction can proceed only on condition that metabolism are in place. Instead, in a view of replication (reproduction) first in the origin of life (e.g. 2, 3), a contrary statement could be expressed as metabolism can proceed only on condition that self-reproduction are in place, because enzymelike functional molecules have to appear by chance again and again if they could not be produced in a self-reproduction system. If it is considered that metabolism does not need functional molecules complicated enough like enzyme but only simple ones, for example, produced in some assumed chemical autocatalysis reactions, then again, the meaning of metabolism needs more explanation in detail. In fact, approving the view of replication (reproduction) first, I agree to the taking out of metabolism from the list, but not because self-reproduction implies metabolism is already in place, instead, because metabolism could be a thing derived from self-reproduction during evolution. Certainly, the sentence the complexity (information) can be considered as product of

Wentao Ma*
College of Life Sciences, Wuhan University, Wuhan 430072, P.R. China

*Corresponding author: Wentao Ma E-mail: mwt@whu.edu.cn

623

624
self-reproduction with change (evolution), on the evolutionary route from simple to complex seems enough to justify the taking out of complexity (information) from the vocabulary list. Based on the definition life is self-reproduction with variations (referred to as definition [2] in the original paper), if variations is considered as a word equivalent to evolution (as the author mentioned), along the logic line above, the final word left seems to be evolution. Apparently, evolution can proceed only on the condition that self-reproduction is already in place. Then, the most concise and conclusive definition should be life is evolution, or more explicitly life is something capable of (Darwinian) evolution, similar to some early suggestions (see 4, 5 for reviews). However, this is apparently not a satisfying definition, because Darwinian evolution is per se a collective concept, only appearing as a concise expression. In other words, we still need to know what kind of things is capable of Darwinian evolution. In fact, seeking a concise expression should not be the only criterion to construct a satisfying definition. A clear definition should catch the most fundamental mechanism underlying the corresponding phenomenon. Therefore, the definition life is self-reproduction with variation should be appropriate but variation does not equal to evolution, instead, it is a feature accompanying self-reproduction which, together with the latter feature, leads to Darwinian evolution. As a further consideration, however, we should notice that this definition remains not satisfying, because self-reproduction is a word typically used in the life science (biology), and should not be used to define life unless we admit that there

Ma
is some special rule for the life phenomenon over other phenomena in nature. In other words, we still need to know how things could self-reproduce in nature. In a recent report of ours (6), self-reproduction (replication) is explained in a chemical background. The main conclusions are: first, self-reproduction (replication) in the substantial world could not mean others but an entity favors the production of its own; second, the major chemical mechanism for such favoring is catalysis, which could be classed into the speed-favoring catalysis and the direction-favoring catalysis (e.g. the enzyme-like function is a speed-favoring one and the template-directing function is a direction-favoring one); third, correspondingly, selfreproduction could be either favoring its own production in speed (self-speeding) or favoring its own production in direction (self-directing), or a combination of them. Let us talk about some scenario in the origin of life to explain these concepts. Life is very likely to have begun with RNA (or RNA-like) molecules (the RNA world hypothesis) (7, 8). Any RNA molecule could direct the chemical reactions of nucleotide-joining to produce its own species via an intermediate complementary sequence, under the mechanism of template-directing copying by base-pairing. That is self-directing. However, only some RNA molecules with characteristic sequences might spread in a pool of random RNAs in the competition for the limited common raw materials, due to their enzyme-like function (so-called ribozyme) favoring the self-directing process. Examples could be replicase ribozymes (catalyzing the template-directed copying) (8), nucleotide synthetase ribozymes (9), etc. That is self-speeding. It is just natural selection that works in this evolution, and would further work on further evolution, selecting for more or higher self-speeding features, upon the prerequisite that different characteristic sequences with these

Figure 1: The scheme of the authors logic (above the dashed line) and my further considerations in this comment (below the dashed line) concerning the definition of life. (A) Constructing the collective (consensus) definition according to the list of common words. (B) Constructing the concise definition by taking out redundant words. (C) Constructing the most concise and conclusive definition along the authors logic. (D) Spreading out the meaning of (Darwinian) evolution to construct a clearer definition. (E) Spelling out the meaning of self-reproduction in a chemical background to construct a definition closer to the essence of life in nature.

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Life: Self-Directing with Unlimited Variability on Self-Speeding

features might appear by variation (in template-direct copying with some errors or other events, such as recombination). In modern life, DNA/RNA is the molecular base of the self-directing and RNA/protein is that of the self-speeding. We could image a bacterium as an entity absorbing raw materials to construct its own offspring (self-directing), in which process a lot of enzymatic reactions occur (selfspeeding). When a variation on the self-directing (originating as the change of its genetic information) appears and if this variation could cause the change of the self-speeding (manifesting as the change of enzymatic efficiency or other phenotypes), Darwinian evolution might take place. If it is emphasized that the Darwinian evolution should be ongoing forever, or say, with an open end (5), the variation of heredity should be unlimited (10), carried by molecules like DNA/RNA or similar polymers, and the corresponding variation of phenotypes should also be unlimited, carried by molecules like RNA/protein or similar polymers. Then the definition should run in such a way as living things are self-directing species with unlimited variability on selfspeeding or more concise life is self-directing with unlimited variability on self-speeding. This definition includes considerations on the chemical base of self-reproduction and

625

implies the most conclusive feature of life, i.e. Darwinian evolution. A scheme of the authors logic and my further considerations in this comment concerning the definition of life is shown in Figure 1. Acknowledgements Financial support by the National Natural Science Foundation of China (No. 30870660, 31170958) and the National 973 Basic Research Program of China (No. 2010CB530500, 2010CB530503) is gratefully acknowledged.
References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). F. A. Anet. Curr Opin Chem Biol 8, 654-659 (2004). A. Pross. Orig Life Evol Biosph 34, 307-321 (2004). P. L. Luisi. Orig Life Evol Biosph 28, 613-622 (1998). K. Ruiz-Mirazo, J. Pereto, and A. Moreno. Orig Life Evol Biosph 34, 323-346 (2004). W. T. Ma, C. W. Yu, W. T. Zhang, P. Zhou, and J. M. Hu. Theory Biosci 130, 119-125 (2011). W. Gilbert. Nature 319, 618 (1986). G. F. Joyce. Nature 418, 214-221 (2002). W. T. Ma, C. W. Yu, W. T. Zhang, and J. M. Hu. RNA 13, 2012-2019 (2007). E. Szathmary and J. M. Smith. J Theor Biol 187, 555-571 (1997).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

Classifying the Properties of Life

http://www.jbsdonline.com In his paper Vocabulary of Definitions of Life suggests a Definition, professor Trifonov (1) analyzes the vocabulary of 123 existing definitions of life in order to provide a path for finding a possible minimal agreement among scientists. To this purpose, he compares from a linguistic point of view the definitions provided in different accounts of life and ranks the terms used therein according to their frequency. He then selects the most used words, which should also reflect the most accepted concepts about life, and groups them under generic concepts representing their common general properties (or meaning). For instance, matter represents the class under which molecules, organic matter and materials fall. Finally, he reduces the conditions that the most frequently occurring definientia represent to the states of affairs or events that require them. In this fashion, for example, energy and material supply are included in the concept of metabolism, while self-reproduction comprises also its essential conditions, metabolism, system, energy and material supply. The outcome of this analysis is a definition of life as self-reproduction with variations. This definition is claimed to be minimalistic for two reasons. First, it is a kind of minimum denominator of the definitions taken into consideration. Second, it is applicable to the minimal structures that can be involved in the origin of life. Moreover, the minimalistic definition is maintained to be generic, as it provides a unique common basis for all varieties of life, including extra-terrestrial life, computer models and abstract forms. However, this proposal raises two crucial questions. Is self-reproduction with changes a good definition? Can this definition actually provide a minimal basis of consensus? One of the Possible Definitions of Life The purpose of a minimalistic definition of life is to reach a minimal consensus. As Trifonov acknowledges (1), there are more than 100 definitions of life, and many of them conflict with each other. The minimalistic definition captures the list of the most recurring defining terms, based on the assumption that recurrence reflects acceptance. A problem with this proposal can arise from the assumptions underlying the grouping of the characteristics and their reduction to generic properties or more complex phenomena. A hypernym semantically includes hyponyms, but this does not correspond to the fact that who accepts a more specific concept is also willing to accept a more generic one, especially when the choice amounts to exclude a distinguishing property. Moreover, more complex states of affairs or events can include more generic concepts or precedent or essential conditions, but causal precedence does not correspond to a logical or an epistemic necessary condition. The effect of this double process of reduction is a definition of life that risks incurring the same problems of the criticized Darwinian definitions (including their chemical and biological variations) (2, 3), characterized by the same characteristics of self-reproduction and variation. As Zhuravlev and Avetisov (4) put it, these characteristics are excessively discriminatory,

Fabrizio Macagno
Instituto de Filosofia da Linguagem (IFL), Faculdade de Cincias Sociais e Humanas, Universidade Nova de Lisboa, Avenida de Berna, 26-C, P 1069-061 Lisboa Portugal

Corresponding author: Fabrizio Macagno E-mail: fabrizio.macagno@fcsh.unl.pt

627

628
as on this perspective it is hardly possible to specify life, including early life, before the emergence of the replication machinery. Moreover, sterile beings such as mules should be excluded from the forms of life (5). Finally, the very concept of self-replication is one of the most controversial matters in the research on the origin of life, or rather minimal life. As Luisi (6) put it:
Even with the simplification of minimal life and way stations, it is clear that the process leading to life is a continuum process, and this makes an unequivocal definition of life very difficult. In fact, there are obviously many places in Oparins pathway where the marker minimal life could arbitrarily be placed: at the level of self-replication; at the stage where self-replication was still accompanied by chemical evolution; at the point in time when proteins and nucleic acids began to interact; when a genetic code was formed, or when the first cell was formed.

Macagno
the properties according to their frequency can be helpful for showing their degree of acceptability. However, such a principle can be combined with other criteria not only hinging on synonymy or causal inclusion. For instance, Kompanichenko (7) started from a similar collection of definitional properties and then selected the fundamental ones based on their discriminating power, or rather their usefulness for the purpose of distinguishing between categories. For instance, accumulation of free energy was chosen because it, better than other properties, allowed one to discern between active and passive systems. This systemic account is different in method and purpose from Trifonovs one, as it is only aimed at providing a possibly shared biological definition of life. However, it points out two methodological dimensions that can be seriously taken into consideration in the classification of definitional properties. The first one is the selection of the kind of definition, and consequently the choice of the concept to be defined (13). The second is the grouping of the properties according to their definitional purpose. The type of definition, or better the definitional sentence, is materially related with the purpose of the definition itself. Ancient dialecticians distinguished between fifteen types of definitions (14), of which the most powerful from a logical point of view was the method of genusdifference. The so-called essential definition was the only one that at the same time guaranteed the convertibility between definiens and definiendum and provided a description of the fundamental semantic properties of the concept defined (9, 15). Modern scientific theories rely on similar types of definitional methods (13), plus the operational (unmentioned in the ancient treaties) and the implicit definition (16-18), which cannot be properly considered as descriptions of the definiendum. All such definitional methods rely on the distinction between the semantic-logical properties of the predicates: the semantic categories (such as substance, having, doing...) and the logical predicables (such as the genus-species or accidental relation between predicates). For instance, life is a noun abstracted from the predicate to continue, to live (19), therefore denoting a state of beings, or a property thereof, but not a substance, matter or entities, which can rather differentiate the biological living state (condition of chemical systems, as stated in some Darwinian definitions, see 6) from computer simulations. Some properties, as the ones used in operational definitions, (3) can be simply accidental, namely can characterize some forms of life, such as life on earth (11). Conclusion The definition of life drawn from the analysis of the properties listed in the existing definitions opens new possibilities of research. One of them consists in applying classical methods of property classification in addition to the frequency criterion. On this view, properties will be grouped according to

The decision of choosing the two characteristics of selfreplication and variation can lead to controversies about the necessary condition of the definition, making it more specific and less general than many approaches to life origin require (7). Moreover, the definition can be also controversial from the point of view of the sufficient conditions. The definition mentions an activity (self-replication) but does not specify any quality that the agent, or rather the logical argument, needs to satisfy. As a result, life simulations can be classified as actual forms of life, as claimed by functionalists who regard life as an abstract process. However, this theoretical position, implicitly supported by the first Darwinian definitions (8), has been strongly disputed (2). The risk with this minimalistic definition is the failure to meet the essential logical requirement of a definition, its convertibility with the definiendum (5, 9), or rather, if we consider other scientific approaches to this logical conditions (4), its being universal and minimal and specific enough. Definitions and Methods of Definition The collection of the properties commonly included in the existing definitions of life and their reduction to a minimal description can be hardly accepted by the different approaches to the problem of life (5). Moreover, the choice of a definition of a controversial concept counts as an implicit support to a specific theory (6), or more generally speaking, a potentially controversial viewpoint (10). This issue becomes much more complicated if we consider that not only are definitions fundamental for finding a theoretical agreement, they are also essential for classifying entities. A genus-difference definition cannot be effective for distinguishing between living and non-living beings in space or on other planets (5, 11), while an operational definition cannot provide a theoretical ground for the origin of life (12). Ranking

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Classifying the Properties of Life

their logic-semantic features characterizing their definitional role. Such properties can be then selected according to their statistical acceptability or hypernymy (or rather genus-species) relations. Moreover, depending on the different purposes of the definition (and definieda) it is possible to formulate different potentially shared minimal definitions. The choice of alternative property classification criteria (for instance, semantic traits or observable properties) can make it possible to provide different statistically minimal kinds of definition, adequate for and aimed at distinct purposes.
References 1. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). 2. C. Cleland and C. Chyba. Does Life Have a Definition?. In Sullivan, W. and Baross J. (Eds.), Planets and Life: The Emerging Science of Astrobiology (119-131) Cambridge: CUP (2007). 3. C. Cleland and C. Chyba. Orig Life Evol Biosph 32, 387-393 (2002). 4. Y. Zhuravlev and V. Avetisov. Biogeosciences 3, 281-291 (2006). 5. 6. 7. 8. 9. 10.

629
J. Oliver and R. Perry. Orig Life Evol Biosph 36, 515-521 (2006). P. Luisi. Orig Life Evol Biosph 28, 613-622 (1998). V. Kompanichenko. Frontier Perspectieves 13, 22-40 (2004). C. Sagan. Life. In Encyclopdia Britannica (2007). Aristotle. Topica. In Ross W. D. (Ed.), The works of Aristotle. Oxford: Oxford University Press (1969). E. Schiappa. Defining Reality. Definitions and the politics of meaning. Carbondale and Edwardsville: Southern Illinois University Press (2003). C. Chyba and C. Phillips. Orig Life Evol Biosph 32, 47-68 (2002). G. Bruylants, K. Bartik and J. Reisse. Orig Life Evol Biosph 40, 137-143 (2010). C. Malaterre. Orig Life Evol Biosph 40, 169-177 (2010). M. Victorini. Liber de Definitionibus. Frankfurt: Peter Lang (1997). F. Macagno and D. Walton. Argumentation 23, 81-107 (2009). P. Bridgman. The Logic of Modern Physics. New York: MacMillan (1927). W. Dubislav. Die Definition. Leipzig: Meiner (1931). H. Dubs. Philosophical Review 52, 566-577 (1943). W. Skeat. The Concise Dictionary of English Etymology. Ware: Wordsworth Editions (1993).

11. 12. 13. 14. 15. 16. 17. 18. 19.

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

Defining Life: Products or Processes?

http://www.jbsdonline.com Trifonovs study of the language used to describe living systems (1) is an eyeopening glimpse into the assumptions underlying much of origins of life research today. Unfortunately, for someone with a putative interest in words, Trifonovs attempt to create categories of words often suffers from a lack of sensitivity to the precise meanings of words, as becomes evident in the meta-categories he invents for analyzing his findings. Evolution and change are not the same thing as any embryologist will tell you. Many things change without evolving. Any physicist would cringe to see force subsumed under energy: The equation f 5 ma has no energy term! These misunderstandings are not, of course, all Trifonovs, but undoubtedly represent fundamental problems in how biologists misunderstand the principles applicable to describing living processes. Such misunderstandings are themselves troubling. Unfortunately, these misunderstandings extend beyond words to fundamental principles as well. Trifonovs definition of life as given by Darwin and Oparin are both incomplete and misrepresent both mens actual theories. Darwinian evolution requires not just reproduction with variations or replication with mutation but also critically! non-random selection. If there is no selection or selection is random, there can be no evolution. Discovering the selection processes operant during prebiotic evolution is an unexplored area of great importance. A second major problem with Trifonovs approach, which to be fair is inherited by him from his sources, is that it focuses on properties instead of processes. The problem of understanding life is not simply describing a system that can replicate and evolve through non-random selection. If these criteria were sufficient, then all human inventions evolve. Indeed, we even have AI systems which employ entities that can acquire computer resources, use them to replicate, change through mutation, and be non-randomly selected. The real problem of understanding life, as Darwin and Oparin understood full well, is how it evolved without a designer or programmer. Thus, the issue of what characterizes the properties of a living system is subservient to the problem of how to evolve such properties through natural processes that existed before any of the properties unique to life had themselves yet evolved. Clearly, whatever this process was, it had to be employ a series of bootstrapping steps in which each new form of organization was able to perform newly emergent functions. Describing and testing such an emergent process is a fundamentally different, and far more difficult, problem than describing the characteristics of living systems that resulted from it (2-4). Indeed, restating the problem as one of processes rather than properties yields a very different set of criteria for what it means to understand life and consequently a very different set of terms for describing it. Some of these criteria are listed in the following table, which I published recently (5). Notably, very few of

Robert Root-Bernstein
Department of Physiology, 2174 BPS, Michigan State University, East Lansing, MI 48824, USA

Corresponding author: Robert Root-Bernstein Phone: 517-884-5039 E-mail: rootbern@msu.edu

631

632

Root-Bernstein

the terms in the Table occur in any of the lists summarized by Trifonov, demonstrating how very different the process approach to understanding life is as compared with the properties approach. The processes approach to understanding life also sheds a very different light on current attempts to engineer living systems. We must remember that even in the building of human edifices and inventions, we often make use of scaffolds that leave no traces on the final products they make possible. Imagine trying to explain how to build a skyscraper without knowing about cranes, bulldozers or cement mixers! Any account of the evolution of life must leave open the possibility, and even the likelihood, that life itself benefited from metaphorical scaffolds that have long been discarded yet were essential to permitting living organizations to emerge. So even if we could manufacture in a laboratory each and every component of a cell from its membranes to its metabolic machinery and its chromosomes, and then mix these engineered components together to produce a functional cell, this would tell us nothing about how life evolved. Even if we keep deleting genes until we reach some apparent minimum without which a cell

cannot function, this tells us nothing about how that minimal set was elaborated, integrated, and selected from the set of molecules and genes that nature elaborated on its random walk toward life. So in defining life, we have to be clear about why we want to define life: is the purpose to be able to make and modify life, or is it to understand how life itself came into existence? Do we want to engineer lifes products or recapture the processes of evolution itself? These are two very distinct questions that will require very different approaches. Unfortunately, too much of recent research in origins of life has confused or even conflated the two. Perhaps Trifonovs study will prompt us to reconsider which questions are of most interest.
References 1. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). 2. R. Root-Bernstein and P. F. Dillon. J Theor Biol 188, 447-479 (1997). 3. A. Hunding, D. Lancet, F. Kepes, A. Minsky, V. Norris, D. Raine, K. Sriram, and R. Root-Bernstein. Bioessays 28(4), 399-412 (2006). 4. V. Norris and R. Root-Bernstein. Int J Mol Sci 10, 2611-2632 (2009). 5. R. S. Root-Bernstein. In: Busiak, V. and Navorro-Gonzalez, R. (Eds.), Astrobiology: From Simple Molecules to Primitive Life, American Scientific Publishers, pp. 285-314 (2010).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

Life?
http://www.jbsdonline.com Life! Needs a definition? Never thought about that. Life! What is it? Been thought about since childhood and surely from times immemorial. Are these two questions same or different? That depends upon semantics. In his paper (1), Trifonov has put forth a definition: Life is self-reproduction with variations. He came to this definition as the minimal essential set through a linguistic as well as mathematical Principal Components Analysis-like analyses of a vocabulary of 123 tabulated definitions of life. This vocabulary depends on the most frequent words used to define life such as life, system, energy, complexity, ability, reproduce among others. This very easily comes to mind that it is the words exactly that have come to Trifonovs aid in defining life. Shabd (Sanskrit/Hindi) 5 Word. Ah! English, not to forget the good old Mathematics role in trimming down the long list here. Does not matter whether the English is written or pronounced. It is said that the primal sound Aum! or Om! is the first word that a new-born child utters immediately after being brought to life in the outside world and a few seconds before the actual crying begins. Life forms other than human beings too have a way of communication with each other and with human beings through sounds; that are words in their context, but sounds for us humans. Nonetheless, a parrot can be taught to speak words while humans can mimic sounds created by these other life forms. The semantics, a context can change a words definition. Words can be considered a life form as well. A word likes to exist within the sphere of one particular definition. Yet, along comes a context trying to change the poor words definition. It is quite possible that because of its resistance to the change from its haven, its changed meaning may be subtle. Take teaching or cheating as an example. There is a variation in the sequence of alphabets in these two words. While in preparatory school, one kid had probably never heard the word cheating. So, whenever classmates would cry out for a child copying in the test, Madam, he is cheating, he is cheating!, the kid would yell along He is teaching, he is teaching!. Well, for one thing, this latter statement is also quite true, because the child is teaching copying to others, of course. And so holds true for DNA sequence variations as well, with the genetic code degeneracy coming to the rescue, leading to a subtle change or even no change at all. The gist is this: Changes do take place. But what endures is the invariant Nature. Inspite of all the fluxes, storms, landslidings, the earth retains its natural green

Seema Mishra
Department of Biochemistry, School of Life Sciences, University of Hyderabad, Hyderabad, A. P. India-500 046

Corresponding author: Seema Mishra E-mail: smsl@uohyd.ernet.in

633

634
charm; the sky still reflects the same blue colour even if clouded by grey clouds once in a while. Inspite of RNAs and proteins being among the few forms of genetic material passed from generation to generation, DNAs continue to rule as the major invariant ones. If life is self-reproduction, then what happens when it becomes a fossil? No possibility of self-reproduction is there. Yet, it still continues to exist in museums. And in the memory. It does not die. Trifonov writes, One unforeseen property of the minimalistic definition is its generality. It can be considered as applicable not just to earthly life but to any forms of life imagination may offer, like extraterrestrial life, alternative chemistry forms, computer models, and abstract forms. This is a nicely philosophical statement. But what about the material things? Material things such as a classical rose-wood rocking chair, arent they beautiful and seeming to have a life of their own? I dont consider material things as life-less, they are also one of the many life-forms. They exist. But what about their ability to self-reproduce, would they be dead simply because they cant? When we undertake research, practically speaking, what is it exactly that we want to do? To understand life and its innumerable little secrets and quirks. If not, then why do research at all? What is it that we are seeking then? As M.Sc students, when we have extracted the DNA from its native place, nothing matches the joy and wonder of seeing the pure, white DNA strands entangled together forming a mesh, just like the web of life. And then we want to know more about how it all forms a beautiful tangled web of life. DNAs and proteins working together in tandem, yet still left undeciphered as to how exactly do they produce a life-form. Trifonov further writes about the Creation of a bacterial cell with chemically synthesized genome, ...However, it should be considered as very much assisted replication as it was provided with an initial natural cytoplasm. And about nearly fully artificial life-form created by Sol Spiegelman, ...This has been an assisted replication as well, since the experiments have been performed in presence of natural replicase.

Mishra
So, without the replicase and the natural cytoplasm, artificial life creations would not have been possible. This means that Trifonovs definition does not entirely fit the bill. And we have to find an invariant notion in the definition of life, if we want life to have an immortal definition. What is that invariant notion then; springing from its source, flowing along and transforming everything that comes in its path and returning to its source in an iterative cycle; which makes life possible in terms of self-reproduction (flow) and variations (transformation)? A beautiful, soulful melody. Let us look outside Science for a clue, a hint, an inspiration. Just like Trifonovs vocabulary list, it is far easier to see that a few words have remained that bind all the many lifeforms across continents, cutting across the barriers. These are Love and Soul. No need to go through complex linguistic and mathematical analyses to reach these two words. Further search for the question that which of these two may represent more rational definition led to this knowledge propounded in Shrimad Bhagvad Gita (2):
Acchedyoyam adahyoyam akledyososya eva cha Nityah sarva-gatah sthanur achaloyam sanatanah!

English Translation follows: Sri Krishna said to Arjuna: This individual Soul is unbreakable and insoluble, and can be neither burned nor dried. The Soul is everlasting, present everywhere, unchangeable, immovable and eternally the same. Can we, then, think of the replicase and the natural cytoplasm above as Soul which makes life come alive? Over to the invariant notion of life, then. Life is Soul!.
References 1. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). 2. Shrimad Bhagwad Gita, Chapter 2, Verse 24.

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

Is A n11 Definition of Life Useful?

http://www.jbsdonline.com In his recent article in this journal, E. Trifonov (1) is in accordance with the seeming consensus among the specialists in the search for the origin of life (chemists, geochemists, biochemists, biologists, exo/astrobiologists, computer scientists, philosophers and historians of science) that there is an obvious need for a definition of life (2). However Trifonov confirms the amazingly high number of definitions of life, leading him to reflect that scepticism is multiplied by the above number, leaving almost no chance for new formulations which, however, continue to appear (1). Then I have two main comments: 1. There is a major issue in the proposed new definition Life is selfreproduction with variations from the analysis of the 123 tabulated definitions of life: Trifonov claims that self-reproduction and changes (evolution) are independent notions but they are not. First, evolution and changes are not synonyms. Variations are not strictly required for natural selection to operate and evolution to occur if there is a collection of various populations of individual systems belonging to distinct lineages that can be formed de novo: natural selection will be able to pick amongst these distinct lineages and thus allow the collection of various populations to evolve. Of course, if sporadic variations are possible in some lineages and if these variations are heritable, then the number of new emergent lineages would be much higher and the evolution much faster. Second, evolution is a process that requires self-reproduction. When considering the above collection of various populations it can evolve by natural selection only if each lineage is self-reproducing. More precisely, as a lineage is basically characterized by (at least ) a set of common properties, the same set of common properties as that of their parents must be reproduced in the descendents in order to maintain the specificity of each lineage and the distinction between lineages and thus allow natural selection to pick amongst lineages (3). 2. More fundamentally, scientifically speaking, I do not think any definition of life useful. I fully agree with some scientists who consider that there has been a dramatic shift in the past years as the question of life is no longer a search for principles of life but has been transformed into a historical issue. The question is no longer What characteristics are found in organisms but not in inanimate objects? but How were these characteristics progressively associated within objects that we call organisms? (4). I agree also that it is always possible that somehow, whether on this planet or another, an abiotic factor with properties totally unknown to us will be able to generate a pattern indicative of evolution of natural selection. However, if such a factor exists, the burden should be to explain why it is abiotic (5).

Marc Tessera
2 Avenue du 11 November 1918, Meudon 92140, France

Corresponding author: Marc Tessera Phone: 133-6-86-46-60-94 E-mail: marc.tessera@wanadoo.fr

635

636
Actually the concept of life is too vague and general, and loaded with a number of historical, traditional, religious values (6). Although life is a useful word in practice, it is not a scientific concept (7). The concept of life is related to an indefinable state. Any definition of life is subjective and arbitrary as is the boundary between living and non-living systems or pinpointing the moment when non living systems would have become living. For instance, saying that virus or prions or vesicles with the capacity of evolving are living systems (or not) adds nothing more than the definition of life one would propose. Finally the statement that any such boundary or moment exists is not falsifiable: no experiment can be considered to prove that it can be wrong. Therefore, as the distinction between living and non living systems is a matter of belief and not science, it is not only hopeless but useless to try to define this indefinable state related to a metaphysical question (3). By contrast the distinction between systems with evolvable capacity and systems without is not so problematic. For example, among known open far-from-equilibrium self-sustained systems there are cyclones. Such systems have a rather complex dynamic organisation which can only be maintained if cyclones are provided continously with matter and energy (i.e. warm steam water vapour) by the local environment (i.e. the sea). The organisation of cyclones is relatively stable during their relatively short span life (only a couple of days on Earth, although the span of life of similar systems can be much longer in some cases: e.g. the red spot of Jupiter which is actually an anticyclone). Such systems did not evolve in a Darwinian sense and the dynamic organisation of such orderly vortex motion has been fundamentally the same since ever. They cannot evolve in particular because they lack the capacity of self-reproduction. There is at least one example of abiotic open far-from-equilibrium self-sustained systems

Tessera
which can self-reproduce: the experimental vesicles with amphiphile bi-layer membranes. These vesicles can grow and bud, a kind of self-reproduction (8-11). However such vesicles cannot evolve because they lack heritable properties independent of the local environment, i.e. independent of the nature of the nutriments provided by the local environment, and thus distinct self-reproducing lineages cannot emerge (3). Only a collection of various self-reproducing lineages of individual systems can be sorted by natural selection and thus can evolve. Actually most of the so-called living systems appears markedly different from inanimate systems because they are the end-products of Darwinian evolution, i.e. a continuum of changes spanning billions of years. There is no point in attempting to define life because of the irreducible metaphysical aspect of the concept. Instead it seems more appropriate to focus on the process of Darwinian evolution as the source of the primordial ancestor on Earth and presumably similar systems elsewhere. The consensus to be reached in the quest for the primordial ancestor must be in defining the minimal process that allows Darwinian evolution to emerge and persist.
References 1. 2. 3. 4. 5. 6. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). S. A. Tsokolov. Astrobiology 9, 401-412 (2009). M. Tessera. Int J Mol Sci 12, 3445-3458 (2011). S. Tirard, M. Morange, and A. Lazcano. Astrobiology 10, 1003-1009 (2010). L. Chao. BioScience 50, 245-250 (2000). P. L Luisi. The Emergence of Life: from Chemical Origins to Synthetic Biology. Cambridge University Press: New York, NY, USA (2006). J. Gayon. Orig Life Evol Biosph 40, 231-244 (2010). P. L. Luisi, P. Stano, S. Rasi, and F. Mavelli. Artificial Life 10, 297-308 (2004). S. Rasi, F. Mavelli, and P. L. Luisi. Orig Life Evol Biosph 34, 215-224 (2004). P. Walde. Orig Life Evol Biosph 36, 109-150 (2006). B. H. Weber. Orig Life Evol Biosph 40, 221-229 (2010).

7. 8. 9. 10. 11.

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

Thermodynamic Inversion and Self-Reproduction with Variations: Integrated View on the Life-Nonlife Border
http://www.jbsdonline.com The excellent work by Trifonov (1) adds to the discussion of extremely important questions: What is life?, or What is the difference between life and nonlife? At present these questions have acquired practical sense, in particular, due to plural experiments on prebiotic chemistry and efforts to obtain an artificial cell. Where is the border between a non-living prebiotic microsystem and the simplest living unit? In the origin-of-life field these both types of systems are often fused into the infinitive term protocell, no border between them. The Trifonovs minimalistic definition Life is self-reproduction with variation, in my opinion, represents a profound insight into the foundation of life. It may serve as an important criterion for the distinguishing an actually alive molecular structure/system, who in the future will inevitably be obtained in laboratory conditions. Some aspects of fundamental biology, Trifonov arises in his article, should be discussed in more detail. Definition of Life or Properties of Life? There exist more than 100 definitions of life. Most of them are true but none is comprehensive. In the opinion of the author of this comments life is such a phenomenon that can not be embraced with an exhaustive definition. Due to this reason, the author in his works (2, 3) deliberately did not give a definition of life. Instead, four key biological properties were formulated on the basis of the comparison of biological and non-biological systems: 1) the ability to concentrate free energy and information; 2) the ability to exhibit an intensified counteraction to external influences; 3) expedient behavior; 4) regular self-renovation on the various levels, including self-reproduction. The essence of the properties could be expressed by the author in the following thesis (or definition): a living system concentrates free energy and information using the ability to exhibit an intensified and expedient reaction to external changes extending its own existence through self-renovation. In fact, the above-mentioned set of four fundamental biological properties provides more opportunities for investigation of the origin of life than the single thesis. The authors goal consisted in consideration of the transition nonliving living systems from different sides; for this purpose the integrative consideration of several properties was preferable in comparison with a single definition. As for the Trifonovs work, apparently, his goal is different: to provide the most compact and objective definition through creative integration of previous efforts in this area. And this goal has been achieved by him.

V. N. Kompanichenko
Institute for Complex Analysis of Regional Problems, 4 Sholom Aleyhem St., Birobidzhan 67016, Russia

Corresponding author: V. N. Kompanichenko E-mail: kompanv@yandex.ru

637

638
Life-Nonlife Border Trifonov emphasizes two independent notions: selfreproduction and changes (evolution); they, actually, exclude one another, as self-reproduction is exact copying, no changes, while changes can not relate to exact copying (1). These combined notions are considered by him as a unique attribute of life, or the border between life and nonlife: all is life that copies itself and changes. According to the authors inversion approach to the origin of life, the decisive step in the transformation of prebiotic microsystems into the simplest living units (probionts) took place due to the thermodynamic inversion, i.e. radical change of the balances free energy contribution/entropy contribution and information contribution/informational entropy contribution into negative entropy values (2-4). The thermodynamic inversion changes the direction of free energy, information and entropy transfer in the interchanging processes between the microsystem and its surroundings: free energy and information start to be imported into the microsystem, while entropy is exported outside. Simultaneously the exchange of substances is reorganized to facilitate extraction from the surrounding energy-rich compounds, providing molecules suitable for constructing new structures in the microsystem. Since the inversion, functional processes arose and connected the main types of biopolymers sequences of nucleotides and proteins. Unlike a polynucleotide or proteinoid chain artificially synthesized in vitro, a functional sequence is meaningful (5). Functional sequences produce function at their destination (binding site or ribosomal translation site). Algorithmic

Kompanichenko
optimization (selection of each symbol in the sequence specifically for function) occurred in the course of internal circulation of bioinformation, through continuous recombination and selection. In this way the inverse prebiotic microsystem becomes an active constituent with respect to the medium. Simultaneously the environment becomes part of the physical medium that is being actively influenced by life. The Trifonovs definition and the authors origin-of-life concept are two different approaches to understanding of distinction between living and non-living systems. In spite of the difference between them, they can be integrated. The Authors Interpretation of the Trifonovs Definition Why are the two opposite attributes, self-reproduction and changes, in the foundation of life, according to the Trifonovs definition? How could they arise? A possible answer can be found in the inversion approach to the origin of life (2-4). The approach offers that thermodynamic inversion might occur only in a prebiotic microsystem oscillating around the bifurcation point under far-from-equilibrium conditions. In the case of balanced oscillations the microsystem acquires a bifurcate structure because of its intermediate position between two attractors the initial and (potential) new states (Figure 1). There appears a paradoxical organization stabilized instability that preceded its transformation into the living unit in the course of thermodynamic inversion. Oscillations of the microsystem around the bifurcation point are not symmetrical: the forward transition over the bifurcation point

Figure 1: Spectrums of the potential states of a chemical system oscillating around the bifurcation point (in the course of nonequilibrium transition from the initial stable state into advanced stable state through the unstable point of bifurcation). A trend to advanced higher-organized state; B trend to advanced lower-organized state; C' and C'' reverse trends to the approximately initial state. On the left (from the bifurcation point) restricting spectrum 1 focused on the initial state, on the right expanding spectrum 2 directed to the set of potential advanced states.

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Thermodynamic Inversion and Self-Reproduction with Variations

brings new accidental change that reflects in the expanding spectrum of the potential advanced states (Figure 1, right part). The back transition is characterized with the restricting spectrum of the potential states because the system strives to return closer to the initial state (Figure 1, left part). Continuous oscillations lead to appearance of new and new changes in the microsystem due to the expanding spectrum of forward transitions; some of them are conserved in congruence with the restricting spectrum of the back transitions. As a result, such an oscillating prebiotic microsystem obtains two opposite tendencies to conservation and modification, at the same time. In the authors opinion, the compact Trifonovs

639

definition of life reflects the consequences of prebiotic processes on the early Earth.
References 1. 2. 3. 4. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). V. N. Kompanichenko. Int J Astrobiol 7, 27-46 (2008). V. N. Kompanichenko. Planet Space Sci 57, 468-476 (2009). V. N. Kompanichenko. In: Seckbach, J. (Ed.), Genesis In The Beginning: Precursors of Life, Chemical Models and Early Biological Evolution (Submitted to Springer, Dordrecht, NL, 2012). 5. D. L. Abel and J. T. Trevors. Theor Biol and Med Model 2:29. doi:10.1186/1742-4682-2-29 (2005).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

Life in Its Uniqueness Remains Difficult to Define in Scientific Terms

http://www.jbsdonline.com Research on lifes genome and proteome and its possible origins is challenging and fascinating. Defining life in scientific terms, however, remains a highly difficult task. Over 100 definitions have been suggested in the course of generations of philosophers and scientists and the list of definitions just continues to grow. This points to an actual lack of a convincing consensus on the definition of life. At the recent 17th Albany Conversation, June 14-18, 2011, New York, E. N. Trifonov presented first results of a unique linguistic word count analysis, performed on the large corpus of all definitions of life. This carefully performed statistical analysis suggested (1) that Life is self-reproduction with variations. In his most recent report in this journal E. N. Trifonov (2) describes the analysis of 123 tabulated definitions of life in high detail. Some of these definitions were also discussed in a special issue of the journal Origins of Life and Evolution of Biospheres (3) reflecting on the difficulties of defining life. E. N. Trifonov argues that most of the definitions do have some shared common sense suggesting that one could arrive to a consensus. Even if the different authors cannot be brought together through space and time E. N. Trifonov organized a sort of voting in absentia by statistically analyzing individual words used in the set of definitions. The words most frequently used might according to Trifonov reflect on their importance, as shared by many of the authors. The careful analysis originated one first attempt for the word count generated definition of life. It goes like Life is [a] metabolizing material informational system with ability of self-reproduction with changes (evolution), which requires energy and suitable environment. This first life-defining attempt was discussed in the context of previous comparative linguistic studies on the definition of life. E. N. Trifonov continued to search for a more concise and shorter definition of life by extracting two terms from the vocabulary of definitions. Trifonov concluded, as mentioned in the context of the Albany Conference above, with the definition of life saying that life is selfreproduction with variations. This definition can indeed be applied as a practical guide in topical origin-of-life research, for example on protocell formation including the encapsulation and elongation of nucleotides (4). Is this the ultimate and universal definition of life? Probably not. Take a selfreproducing malicious software script like a computer virus, a computer worm, or a Trojan horse. Computer viruses spread from one computer to another via a network or via internet by reproducing themselves. After several hundreds of copies (and infected computers) random variations will provoke that the latestgeneration offspring virus is not to 100% identical to the ancestor virus script. A computer virus performs self-reproduction with variations. It is not alive. Please note that variations observed in biological evolution such as mutations are

Uwe J. Meierhenrich
ICN, UMR 6001 CRNS, University of Nice-Sophia Antipolis, 06108 Nice, France

Corresponding author: Uwe J. Meierhenrich Phone: 133 (0) 492 076177 Fax: 133 (0) 492 076151 E-mail: Uwe.Meierhenrich@unice.fr

641

642

Meierhenrich

Figure 1: Wordle assembled keywords on the definitions of life.

non-directed variations. They are sudden, spontaneous, and random changes (5, 6) that do not imply any driving force other than dissipating energy. What about other definitions of life? (Figure 1) Wikipedia, for example, posts that Life is a characteristic that distinguishes objects that have self-sustaining biological processes from those which do not. The difficulty with this definition is that it refers to biological processes. Biology as explained by Wikipedia is defined as the study of life. Here we turn in a circle. This is elegant but not helpful. Such as Your right is where your thumb is left. I do not attempt to give a competing definition of life here. Being physico-chemist I would prefer definitions of life that incorporate lifes performance and control of metabolism, including autocatalysis, cyclic kinetic processes, feedback loops, and active transport. Because such definitions may better help to detect and characterize life in extraterrestrial and extrasolar planetary systems. Interestingly, the linguistic analysis applied by E. N. Trifonov for the definition of life is conceptually close to a method known as Principal Component Analysis (PCA). PCA is a mathematical procedure extracting the major (principal) components covering most of the data for further statistical treatment and analysis. As outlined by E. N. Trifonov (2) PCA has been applied by himself (ENT) to decipher a consensus

on the temporal recruitment order of amino acids into early proteins (7, 8). Further work should correlate the important PCA-derived recruitment order of amino acids with experimental data on the formation of amino acids for example in the prebiotic atmosphere (9) and even in interstellar environments (10, 11).
References 1. E. N. Trifonov. J Biomol Struct Dyn 28, 1060 (2011). 2. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). 3. J. Gayon, C. Malaterre, M. Morange, F. Raulin-Cerceau, and S. Tirard (guest Eds.), Special Issue: Definitions of Life. Origins Life Evol Biospheres 40, 119-244 (2010). 4. U. J. Meierhenrich, J.-J. Filippi, C. Meinert, P. Vierling, and J. P. Dworkin. Angew Chem Int Ed 49, 3738-3750 (2010). 5. P. Atkins. On Being: A Scientists Exploration of the Great Questions of Existence. Oxford University Press (2011). 6. U. J. Meierhenrich. Angew Chem Int Ed 50, 9240 (2011). 7. E. N. Trifonov. Gene 261, 139-151 (2000). 8. E. N. Trifonov. J Biomol Struct Dyn 22, 1-11 (2004). 9. A. P. Johnson, H. J. Cleaves, J. P. Dworkin, D. P. Glavin, A. Lascano, and J. L. Bada. Science 322, 404 (2008). 10. G. M. Muoz Caro, U. J. Meierhenrich, W. A. Schutte, B. Barbier, A. Arcones Segovia, H. Rosenbauer, W. H.-P. Thiemann, A. Brack, and J. M. Greenberg. Nature 416, 403-406 (2002). 11. C. Meinert, P. de Marcellus, L. Le Sergeant dHendecourt, L. Nahon, N. J. Jones, S. V. Hoffmann, J. H. Bredehft, and U. J. Meierhenrich. Physics of Life Reviews 8, 307-330 (2011).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

Definition by Means of Indefiniteness

http://www.jbsdonline.com Humans recognize themselves as a part of humanity, of the biological world and of life simultaneously. They identify themselves differently in each case. However, in the absence of an accurate definition of life, the process of the identification of humanity remains incomplete. The incompleteness of this process justifies the numerous attempts to define life. Nevertheless, none of these previous attempts has been indisputably successful. This previous lack of success implies that life possesses some elusive feature(s) escaping the (current) definition. In a recent paper (1), Edward N. Trifonov tries to find these features of life with a word count approach. This work is akin to publications that analyze sets of published definitions (see reference in ref. 1), but it differs from them in its level of generalization. The definition obtained after such generalizations, is, I am afraid, only loosely connected with the essence of life. Four comments below will illustrate the reasons for my mistrust. 1. Concerning the approach. Any statistical analysis is fruitful if and only if the sample is representative. In this case, there are two uncertainties: whether the definitions for analysis were selected correctly and whether the selected definitions actually include the principal features of life. Both aspects are of importance because statistics cannot help to discover characteristics absent from the sample. As I noticed above, the life possesses some elusive feature(s) escaping the (current) definition so life cannot be defined exhaustively at present time. Statistical analysis (even if the second aspect of the problem, i.e., the content of the definitions, is incomplete) can be useful for studying the history of scientific reflections about life. However, the latter is not the purpose of the work reviewed here. The goal of the work is to give a possible shorter definition of life containing components that are both necessary and sufficient. To achieve that goal, all the words from the 123 definitions of life published at different times were considered, and the frequency of these words was calculated (Table 1). The relative frequency of use of the words was not considered in the subsequent analysis. Instead, the words were combined in 10 groups according to their common meaning, and the 6th group was given further consideration. The author was clearly under the impression that the 6th group was chosen on the basis of word count. However, this group was clearly chosen on the basis of an evaluation of the inclusive capacity (although no measure of capacity was presented in the text) of certain arbitrarily selected words. All of the subsequent analyses in the paper are unrelated to word counts. 2. Concerning the subject under consideration. In discussions of the definition of life, many authors use (explicitly or otherwise) the terms life and

Yuri N. Zhuravlev
Institute of Biology and Soil Science, Russian Academy of Sciences, Far Eastern Branch, 159 100-letiya Ave, 690022, Vladivostok, Russia

Corresponding author: Yuri N. Zhuravlev E-mail: zhuravlev@ibss.dvo.ru

643

644
organism as synonyms (2). This shortcoming can be discovered in many of the 123 papers reviewed. Strictly speaking, this shortcoming makes the exploitation of the method of principal components unacceptable in this case. This shortcoming explains (in part) the reason that the definitions are more than often in conflict with one another (because different subjects were treated). In the paper reviewed, the author avoids this topic entirely. However, to disregard the distinction between life and organism is a shortcoming of great significance. To interpret life as an organism means to overvalue the genetic constituents of the system and to underestimate all other features. I hypothesize that a biological object cannot be exhaustively defined solely in terms of its genetic constituents. It is necessary to combine the internal and external definitions of a biological object (3). The internal definition is primarily genetic and considers the biological object as a triad: Oint 5(P, F, Ph), where p denotes program, f functions and ph observables [see details in (3)]. The successions, recursions and compositions of mappings (on the basis of these three components) make the object. The external definition reflects the position and role of the biological object in its surroundings (in unity of living and non-living). Here, the object can be interpreted as a certain operator converting the surroundings: Oext5F:S1S2. This definition is more ecological. Separately, no definition can be sufficient to define a biological object. However, both definitions, even if taken together, are insufficient to define life as a system because the multitude of biological objects is only a list of elements related to life, whereas the production of the system from its elements introduces novel properties that cannot be inferred directly from the characteristics of the elements. Thus, the definition of life as a system must include characteristics absent from the definitions of biological objects. The potential infinity of life (in contrast to the finite nature of objects) is an example of such novel characteristics. Again, this topic was not addressed by the paper and that remains unclear what is characterized through exact replication. 3. A comment regarding the attempt to improve Darwin. The short Darwinian formula descent with modifications, which Darwin used to ground the theory of origin of species (not the origin of life), was transformed by the author into self-reproduction with variations. Even if one pays no attention to the legitimacy of such a transformation, a question remains: are the pairwise substitutions reproduction/descent and variation/

Zhuravlev
modification (as a substitution of one versatile term for another such term, where the content of the two terms overlaps in part) fruitful in any sense? Here, the author uses the term self-reproduction, which he explains as exact replication of the ideal RNA duplex , thus obtaining the definition that life is exact replication with variation (non-exact). Generally speaking, the neologism self-reproduction lies beyond intuition and even beyond second law of thermodynamics, but its content (used in the paper) was not specified by the author. As result, clear Darwinian formula was converted in the vague vulnerable allegation in which nothing of Darwinian formula was conserved. 4. The title of the paper, Vocabulary of Definitions of Life Suggests a Definition, gives the impression that the approach used was valid (almost without the authors participation) to produce the conclusion that all is life that copies itself and changes. I believe that the author intervened in at least three principal instances: first, through the supposition (undoubtedly incorrect) that the definitions considered include all necessary and sufficient properties of life; second, through the substitution of the word-count method for a method of inclusive capacity evaluation; and third, through the transformation of the Darwinian formula [operating with natural selection and organic beings (4)] into the definition of life [operating (here) with undefined subjects] and through subsequent generalization. This tangled and dashed line, by which the author connected the word count results with the meaningful laboratory experiments by Spigelman, can be characterized as an operation of arbitrary treatment but not that of compression whereas the latter operation is usually used in information theory to convert a string in the shorter string (5). As consequence of this arbitrary treatment, the image of the soap bubble divided into two smallest bubbles is easily located in the scope of the final definition of life (as it is given in the paper reviewed), but this waning image is inappropriate to recover the plethoric image of the biological object or of that of life as a system.
References 1. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). 2. Yu. N. Zhuravlev and V. A. Avetisov. Biogeosciences 3, 281-291 (2006). 3. Yu. N. Zhuravlev and V. A. Avetisov. In: Genetic Transformation, M. Alvarez (Ed.), InTech, 29-52 (2011). 4. C. Darwin. Origin of species, John Murray, London (1859). 5. P. M. B. Vitanyi and M. Li. IEEE Trans on Inf Theory 46, 446-464 (2000).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Comment

The Definition of Life and the Life of a Definition

http://www.jbsdonline.com The meaning of life is probably of interest to any reflective individual. However, the definition of life is of interest to a smaller circle of researchers trying to better elucidate the meaning of life in a reflective, critical, and constructive manner. To critically evaluate a novel definition of life one has to take a step backward and understand the meaning of definition. In this context, I would like to draw what I consider an interesting analogy between our understanding of measurement and our understanding of definition. At the beginning of human civilization, measurement took the form of a one-to-one correspondence between natural numbers and pre-defined objects. Therefore, measurement was identified with the counting of objects that exist, and I emphasize this point, prior to the measurement/counting procedure. For instance, the ancient man could have counted his wives by using his fingers in the same way as young children learn to count objects. However, modern science has turned the idea of measurement on its head in the sense that measurement is now understood as the operation through which we define objects whose existence cannot be trivially assured prior to the measurement process. For instance, potential energy does not exist in a similar manner to fingers or sheep. Potential energy as an object of scientific inquiry is operationally defined, for instance, through the procedure of multiplying m, h, and g. Our understanding of definition has been changed along the same line. Let me explain this argument. Alchemy was obsessively involved with purifying the spiritual essence of matter the same as Plato was obsessed with purifying the essence of beings through definitions. In retrospect, it seems that science has advanced not by purifying the essence of objects, whether natural or conceptual, but by adopting the idea of measurement/definition as (1) producing a novel object of contemplation (e.g., potential energy, imaginary numbers) and (2) pragmatically proving the benefits of this new measurement/definition. With this context in mind, I would like to discuss Trifonovs interesting paper (1). The paper is a highly intelligent attempt to define life by purifying its conceptual gist from a corpus of definitions. This methodology is in line with current attempts to harvest collective intelligence (2) for understanding the meaning of a concept. In this case, the collective intelligence is that of the experts who defined life. Revealing the most frequent terms in the definitions is a right step in this direction. However, the frequency of terms may represent their base-rate in language rather than their informational value in the context of the definitions of life! This is the reason why raw frequencies are rarely used in automatic text processing but other, weighted measures are used (3). Clustering the words composing the definitions under different titles is not a trivial task, as the criterion for clustering and the justification for using it should be stated. Trifonov does not use a structured method for clustering but intuition only. Comparing his results to those of Kompanichenko is of minor use in establishing the validity

Yair Neuman
Department of Education, Ben-Gurion University of the Negev, Beer-Sheva, Israel, 84105

Corresponding author: Yair Neuman E-mail: yneuman@bgu.ac.il

645

646
of the clustering procedure, as Kompanichenkos data are not a gold-standard and do not follow a strict clustering procedure. In this sense, comparing the intuitions of two researchers is of minor value for scientific validation as their intuitions may converge while having nothing to do with the real state of affairs. I must emphasize that I have nothing against intuition, but one has to remember that the clustering procedure used in this paper cannot pretend to have validity beyond intuition. Moreover, Trifonovs attempt to seek components that are both necessary and sufficient cannot be accomplished through the clustering procedure. Cluster analysis, in its statistical sense and common scientific use, does not support the identification of necessary and sufficient features through which we may identify the essence of a concept. In addition, we should remember that Wittgenstein has already shown that different instances of a given concept do not share a fixed set of features that may be analytically used for defining it. To recall, Wittgenstein coined the term family resemblance (4) as a part of his attack on essentialism. The idea behind family resemblance is that instances of a given concept (e.g., Life) are united not by a single common defining feature, but by a complex network of overlapping and criss-crossing similarities. (5). For instance, living systems, from the brain to the immune system and the fungi may be generally characterized as meaning making systems (6). However, the different expressions of meaning making do not share the same features but criss-crossing similarities. Based on the above critique, Trifonovs justifications for the process of clustering and extracting the components raise some questions. For instance, why complexity (information) is a product of self-reproduction with changes (evolution)? Is self-reproduction with changes the cause of complexity? In what sense? Lets take a simple case in which a string S1 comprised of letters from a finite predefined alphabet is transformed through copying + change into another string, S2. It is trivial that for the long run, randomly shuffling S1 would result in the increase of entropy that might be mistakenly interpreted as increased complexity. If the complexity of the

Neuman
transformed string is genuine, what is the precise meaning of the change through which the algorithmic complexity of the string, for instance, increases? Can we argue that the meaning of this complexity is necessary for defining the meaning of the change and therefore complexity is not the product of change but rather a necessary term for defining it? These questions and many more pop up while reflectively contemplating Table II in ref. 1. The final definition of life self-reproduction with variations is economical and impressive, and Trifonov concludes by arguing that the definition is naturally required for the exploration and that it is not a purely philosophical historical matter. Researchers in biology are well familiar with the ideas of self-reproduction and variation, and I wonder whether combining these components in a minimalist definition of life produces a synergetic effect. Lets return to the example of potential energy. While h, m, and g are three trivial objects, their multiplication synergistically creates a new entity that can be further developed (e.g., elastic potential energy, nuclear potential energy) and used for measuring and understanding the physical world. How can we use Trifonovs novel definition of life in guiding us toward a better understanding of life? Answering this question is a must in evaluating the potential benefits of the definition. Let me conclude my commentary by complimenting Prof. Trifonov for addressing a fundamental question in life sciences. Hopefully the questions raised in my commentary may catalyze our understanding and the evolution of his novel definition.
References 1. E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). 2. P. Levy. Collective intelligence. New York: Basic Books (1997). 3. C. D. Manning and H. Shutze. Foundations of statistical natural language processing. Cambridge, MA: The MIT Press (1999). 4. L. Wittgenstein. Philosophical Investigations. Oxford: Blackwell, (1953). 5. H.-J. Glock. A Wittgenstein Dictionary. Oxford: Blackwell (1996). 6. Y. Neuman. Reviving the living: Meaning making in living systems. Oxford: Elsevier.

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Journal of Biomolecular Structure & Dynamics, ISSN 0739-1102 Volume 29, Issue Number 4, February 2012 Adenine Press (2012)

Author Response

Definition of Life: Navigation through Uncertainties

http://www.jbsdonline.com I am pleased to receive such multifaceted response (1-19) to my paper (20). It helped me to better realize what exactly I have done. Perhaps, the main thing is original motivation: how from 123 uncertain definitions of the uncertain phenomenon described in uncertain terms to derive a consensus, without engaging in the debates, which so far did not bring the consensus. As P. L. Luisi put it: the concept of life is too vague and general, and loaded with a number of historical, traditional, religious values (21). The debates, therefore, have been intentionally excluded from my analysis. No semantics, logics, semiotics, and alike, nor philosophy in general were involved. With all respect to philosophy, mother of sciences, I chose to keep away from it, with the risk of becoming non-scientific, and engaged in the word-count approach, vocabulary method instead of insight (4), which has never been tried for definition of life. The first consequence, of course, is an understandable avalanche of protests, but a comparable flow of praise as well. Thus, the main motivation and the main point of the disputed paper was to bypass centuries-long philosophical debates on the definition of life, lacking the cohesiveness (13), which, as I see it, continue to lead nowhere, and suggest an entirely new approach on a new ground, well away from the old territory. This point is not appreciated by most of the comments dragging instead back to the weathered grounds (1, 2, 4, 6-16, 18, 19). The risk with this minimalistic definition is the failure to meet the essential logical requirement of a definition (8). Yes, indeed, as it was not geared to the traditional routines of definitions. Another intention in deriving the minimalistic definition was to find, hopefully, a practical guide towards potential minimalistic models of life. The resulting threeword definition is considered by many as incomplete. A whole variety of definientia to supplement it is offered: heritable variations (3), information, energy, environment, thermodynamic inversion (5), error threshold (6), self-directing and self-speeding (7), exchange with the environment, kinetics and self-assembly (11), cell (11, 15, 18) (I do not consider cell as unit of life, see below), adaptive evolution (13), selection (14, 18), metabolism (16), lack of purpose, evolvability (18), and Love and Soul (10). The last one deserves special comment. Following Cartesian body/soul division I focused, as many others before, at the body (structure, mechanisms). The soul, as well as mind, consciousness, love remain firmly in the philosophical and theological realm. Apart from additional defining words full alternative definitions are suggested as well (e.g., 3, 4, 10). Accepting the above suggestions would completely sterilize and smear the original idea of the paper. A fable of S. Mikhalkov Elephant-painter suits here as ironical metaphor. The elephants landscape painting was criticized by other

Edward N. Trifonov1,2
1

Genome Diversity Center, Institute of

Evolution, University of Haifa, Mount Carmel, Haifa 31905, Israel

2

Central European Institute of

Technology and Faculty of Science, Masaryk University, Kamenice 5, CZ-62500 Brno, Czech Republic

Corresponding author: Edward N. Trifonov Phone/Fax: 1972 4 828 8096 E-mail: trifonov@research.haifa.ac.il

647

648
animals for lacking Nile, and snow, and kitchen-garden (22). The spectrum of the suggested additions to the definition also vividly illustrates the starting point of the paper: derivation of the consensus definition of life by the way of traditional disputes leads only to further inflation of the definitions and to accumulation of disagreements. The minimized definition fails to illustrate the myriad of possibilities of lifes emergence (16). The minimized definition is not to illustrate, but to suggest what is common for that myriad. Another comment is the question What is Life? hardly can be considered scientific. Falsification is impossible (6). But would not we still try to imitate life as close to its essence as possible? Why should we surrender to Popperian bounds, if current working hypotheses continue bring fruits of new knowledge? The philosophical disputes are often about terminology at the expense of essence. Several comments are actually, terminological (which term is better to use): Description instead of definition (1), evolution instead of variation (7), processes instead of properties (are not self-reproduction and mutation both processes?), understanding instead of definition (14), and other. Is it, really, important how exactly one or another thing is called when a simple (nave some would say) common sense picture is to be drawn? Few self-explanatory words are put together, describing what would be the target in the search for minimal system/process/network/transition at the border between life and non-life. The enchanting exercise (2) of word count is meant as the way out of terminological multinode net, to the simplest what to look for. The recipe, whether right or wrong, should be minimalistic, and one such recipe is offered. I hope that the definition suggested will be useful in the search for the border, and I am glad that this hope is shared, though not by all (see below). I did abandon the rival grounds after suggesting the nine definientia. They may serve as, again, a tentative minimal set of relevant terms (notions, categories) to continue the debates, perhaps, on more fruitful basis. I did so with some curtness (7) since, after all it is not exactly my territory. The suggested minimal definition is, obviously, debatable, and the final assertion of the definition of life needs more cautious and deeper consideration (7). One possible outcome is the construction with self-directing and self-speeding (ibid). I suspect, however, that the debates would not go far away from the minimalistic definition (if only one-two words are added). Lack of sensitivity to the precise meanings of words (14) is common criticism. The 123 definitions are all fuzzy (2) in various degrees. They emanate from enormity of the problem, and belief that only humble descriptions (1) may be suggested. The vocabulary of the definitions is fuzzy as well,

Trifonov
often giving different names for the same thing, not mentioning the eternal disagreements what would be the meaning of this or another word. The attempt to classify the words, as in the disputed word-count paper, may only be fuzzy as well as the concluding definition., as it is, indeed (7, 19), The word force is good example (14). Strictly, it is not energy, however, the only word group it may belong to is Energy. After all, force is dv/dt, and v2 is energy. Another example is uncertain meaning of exact replication (19). Again, it may belong, obviously, to Reproduction, rather than to any other of the nine groups, irrespective of what exact meaning would be given to it. Evolution and changes are not synonyms (17). Yes, but the suggested groups of words with similar meanings are not groups of synonyms. The clear Darwins formula descent with modification (19) is as fuzzy as selfreproduction with variations. Yet one more example of different understanding is derivative nature of complexity: it is asked by one commentator why complexity (information) is a product of self-reproduction with changes (evolution)? (12), while according to another comment (7) it goes without question: certainly, the sentence the complexity (information) can be considered as product of self-reproduction with change (evolution), on the evolutionary route from simple to complex seems enough to justify the taking out of complexity (information) from the vocabulary list. Viruses are in the strictest sense incapable of self -reproduction (16). But the strictest sense is avoided intentionally, otherwise none of the words of the Reproduction group will go together. And no groups will be formed at all. One argument against the alleged redundancy of the minimalistic definition is that error-free replication (more precisely, any information transmission process) is impossible (6). This is clear case of misunderstanding: the variation in the definition is meant, of course, as inherited propagated variation, not just error, that is mostly lethal. Classification of definitional properties (8) is a whole universe of uncertainties. Every single classifying word would invite disputes. For example, does metabolism belong to Chemistry, or to Life, or to System? The most frequent words of the vocabulary are life and living. Strictly speaking, they do not belong to the same group of meanings, though common sense (or intuition) would put them together. How to classify the words without coming to absurd extremes of a definition (all inclusive definition, or meaningless one-two word stumps, like living matter, or system and environment)? By suggesting those nine groups of related words I have taken a risk to find a golden middle that suits my intuition and common sense. And I do have reasons to believe that it is close to the intuition and common sense of many. The end result of the anthropomorphic consensus polling(2) nine or so word groups that could serve as definientia is,

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

Definition of Life: Navigation through Uncertainties

essentially, there, no matter how accurately the groups are gathered. Excessive sensitivity to precise meanings would end, perhaps, in up to hundreds of word groups, to completely blur the target. Is the definition so loosly constructed, vulgar for scholastic perception, even non-scientific (e.g., 9, 16, 17), useful in any way? Opinions divided. There are few on the positive side (e.g. 1, 5, 6, 9, 12), while many disagree (2, 15-17). Does that definition have a euristic power (6), and what it is useful for? One opinion is the distinction between living and non living systems is a matter of belief and not science, it is not only hopeless but useless to try to define this indefinable state (17). And How can we use Trifonovs novel definition of life in guiding us toward a better understanding of life? (12). I have not seen that efforts to define life have contributed at all to that understanding (J. Szostak, 15). Yet, the Szostaks definition self-sustained chemical system capable of undergoing Darwinian evolution is broadly quoted. It, thus, has something in it that appeals to researchers of life. Is not that a manifestation of some better understanding? My own definition helped me to realize, for example, that cell, probably, would not be needed as part of minimal definition. Thus, the efforts to imitate the minimalistic life, perhaps, do not have to include the attempts to build the cell, as, say, in the work of Szostak (23). As it is put correctly in (14), we have to be clear about why we want to define life: is the purpose to be able to make and modify life, or is it to understand how life itself came into existence? I thought of pursuing both targets. In the process of construction of the minimalistic life, hopefully, guided by the minimalistic definition, one certainly will arrive to better understanding. The definition quite likely has potential to yield genuine biological insights (6). This definition can indeed be applied as a practical guide in topical origin-of-life research (9). An immediate concrete example is recent experiment with synthesis of Gn on template of GCCn, checking whether G would be occasionally incorporated opposite G as well, thus, evaluating possibility of mistakes in the presumably ancient replication system (24). This system based on GCCn has been suggested as the minimal process (17) in (20 and references therein). The mentioned work is part of an effort to design minimalistic system (process) in accordance with the minimalistic definition. Questioning the usefulness of the definition inevitably puts the whole work under question. Indeed, some comments are firmly negative (14, 19). Some, however, consider it as an important contribution, resolutely so (1, 4, 5, 7, 12), or reluctantly (14). At the same time a frequent motif is that the definition of life is simply impossible (1, 5, 15, 19). The methodology of the word-count work is, generally, accepted with interest. The extreme negatives are given by The ranking of words according to frequencies seems blind

649

to the underlying logical relationships (4), and There is no genuine scientific justification behind this approach and no guarantee that the numerous compared definitions are not all based on common misconceptions (6). Yet it is delightfully clever, objective and quantitative approach to defining life (3), and sound analytical effort applied rigorously on a comprehensive body of literature (11). More moderate criticism relates rather to suggestions on improvement: Trifonov should not stop at the very first principal component of his statistical vocabulary filtering approach (2). This suggestion is, unfortunately, unrealistic as the statistical ensemble for the possible second component would be too small. Supplementing the list of definitions by data from other sources would be justified (3) but these data have to be not single individual definitions, like the one by V. Kunin (25), as suggested in (13). I operated with known collections of definitions, which I did not compile myself, not to get excessively biased. Property classification should have been performed (8). Weighted measures of information capacity of various words should be used, and structured method for clustering rather than intuition only (12). Inclusive capacity of the groups has not been estimated (19). Attempt to seek components that are both necessary and sufficient cannot be accomplished through the clustering procedure (12). At this point I would disagree, since the clustering was not used for that purpose in the paper. Rather some straightforward (non-scientific!) intuitive reasoning. The above suggestions are acknowledged and will be considered in future work. Several cases of the Is that life? category popped up in the discussions. Frost tracery on a window pane or frostwork-type mineralizations in cave deposits (2) is suggested as a non-life example that fits to the definition. I would not agree with this, since the replication of ice crystals shows variety of shapes, but the same variety that does not change. Each crystal type reappears unchanged. The indivisible pair selfreproduction/variation, of course, implies that the variation is copied in the next reproduction cycle. Thus, the statement that self-reproduction without variation would be entirely fictitious, in that it can never be realized as a natural process (2), is trivial. It simply says that every copying is non-exact. But is the copying mistake always inherited? Another example of the confusing non-life phenomenon with life is the soap bubble divided into two smallest bubbles (19). There is no variation component in it, and it is division, rather than replication. Sterilized cat and frozen bacterium (4), as well as mules (8, 11) are just manifestations of life, or aberrant forms of life not a challenge to any general definition. The fossil that continues to exist in museums and in the memory does not die (10). But it does not live either, as it does not make copies. On the other hand, a computer virus performs self-reproduction

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012

650
with variations. It is not alive (9). By my definition it is. I thought that the two key features are applicable not just to earthly life but to any forms of life imagination may offer, like extraterrestrial life, alternative chemistry forms, computer models, and abstract forms (20). Similarly, considering the generality of the definition it is wrong to state that I pound on the RNA-world drum (13). In conclusion, after reading the comments I realized that although what I have done is not in the main stream of research on definitions of life, and not fully justified methodologically, the result, as fuzzy as it is, and, thus, questionable for some (but complimented by others) gives the tentative self-explanatory concise answer to the questions what is that we are all looking for, and what to do to get it, although exact wording may need some brushing to become academically approved. The reader interested in the subject of defining life and explaining its early evolution will find sufficient substance in (20) to make this article worth reading and instructive (13). I am grateful to all commentators, who joined my humble efforts to move towards elusive target origin of life for many suggestions and thoughts to contemplate. I hope to continue the discussions beyond this time- and space-wise brief exchange, as many questions remain unanswered.
References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21.

Trifonov

22. 23. 24. 25.

E. Di Mauro. J Biomol Struct Dyn, this issue R. Egel. J Biomol Struct Dyn, this issue H. G. Hansma. J Biomol Struct Dyn, this issue G. A. J. M. Jagers op Akkerhuis. J Biomol Struct Dyn, this issue V. N. Kompanichenko. J Biomol Struct Dyn, this issue E. V. Koonin. J Biomol Struct Dyn, this issue W. T. Ma. J Biomol Struct Dyn, this issue F. Macagno. J Biomol Struct Dyn, this issue U. J. Meierhenrich. J Biomol Struct Dyn, this issue S. Mishra. J Biomol Struct Dyn, this issue A. Mittal. J Biomol Struct Dyn, this issue Y. Neuman. J Biomol Struct Dyn, this issue R. Popa. J Biomol Struct Dyn, this issue R. Root-Bernstein. J Biomol Struct Dyn, this issue J. W. Szostak. J Biomol Struct Dyn, this issue B. L. Tang. J Biomol Struct Dyn, this issue M. Tessera. J Biomol Struct Dyn, this issue F. M. Yeong. J Biomol Struct Dyn, this issue Y. N. Zhuravlev. J Biomol Struct Dyn, this issue E. N. Trifonov. J Biomol Struct Dyn 29, 259-266 (2011). P. L. Luisi. The Emergence of Life: from Chemical Origins to Synthetic Biology. Cambridge University Press: New York, NY, USA (2006). http://www.youtube.com/watch?v=q44T40nuAAY S. S. Mansy, J. P. Schrum, M. Krishnamurthy, S. Tob, D. A. Treco, and J. W. Szostak. Nature 454, 122-125 (2008). S. Pino, E. N. Trifonov, and E. Di Mauro. Genomics Proteomics Bioinformatics 9, 7-14 (2011). V. Kunin. Orig Live Evol Biosph 30, 459-466 (2000).

Journal of Biomolecular Structure & Dynamics, Volume 29, Number 4, February 2012