PART-1 Comprehensive Exam- June 2013

Part 1 (SECTION A) 9AM 1PM

Instructions: Attempt 8 questions in this section. You have to pass 6 (six). Please spend
no more than 30 minutes on each question.
Begin the answer to each question on a fresh page.
Q1
Which of the following is more critical for the survival of the organism (i) low error rate
during transcription (ii) low error rate during translation or (iii) low error rate during
replication. Describe the reason behind your answer.
The error rate of translation in prokaryotes is 1 wrong amino acid in every 1000 to 10000
amino acids incorporated into protein. List two strategies that might be utilized by the
translation apparatus to ensure high fidelity of incorporation. Describe an experiment to
test each listed strategy.

Q2
Next-generation sequencing technologies provide significantly cheaper and faster
alternatives to sequencing genomes. These technologies typically produce a large number
of short reads, making these approaches semi-quantitative. Assume that a transposable
element is acquired by a bacterium and this is integrated into the chromosome. In theory,
the transposon should integrate at some random position in the chromosome. However, in
the cell, there may be parameters which make transposon integration more likely at
certain loci than another.
a.
b.

Discuss what these parameters might be.

Using next generation sequencing, devise an experiment that detects hotspots, if
any, of integration of horizontally acquired DNA in bacteria.

Q3
Sketch out the long-term behaviour of all the solutions to the following dynamical
system:
= ! + ! 1
= ( )
Your answer should include all fixed points, and saddle separatrices if any. Also write
down in words the nature of each fixed point. [You should start with nullclines to break
up the plane into different regions, and try to find the direction of flow within each
region. This should be sufficient to guess what all trajectories look like. You may
linearize the system if you wish but it is not necessary.]

Q4
The allele for curly hair and the allele for straight hair show incomplete dominance. The
heterozygote has wavy hair. The Punnett square below might be useful.
a. Consider the following two populations.
Population A: In this population, the f(straight hair) = .36, f(wavy hair) = .48, and
f(curly hair) = .16.
Population B: In this population, the f(straight hair) = .16, f(wavy hair) = .68, and
f(curly hair) = .16.
You do not know what the breeding patterns have been. For each of these
populations, determine (a) what the allelic frequencies are for this gene and (b)
whether the population is in H-W equilibrium.
b. After a few generations of random mating for all populations in question,
Population A consists of 250 individuals. Fifty members of Population B (a
random selection of the members of that population) pack up and move into
Population As territory. The new members mix and mingle with the old
members, and they become a single population, freely interbreeding and mating
randomly. After a few generations of this random interaction, what would be the
new frequency of wavy haired individuals in the hybrid population?
c. Back in the original Population B, which has happily continued randomly
breeding, despite the departure of 50 individuals, a strange plague strikes the
population and kills every individual who has curly hair. Through grief-stricken at
their loss, the remaining population continues its random breeding activities. After
one generation of random breeding after the loss, what would be the new f(curly
haired) among the new babies?
d. Sadly, the plague has apparently not burned itself out, and it wipes out a second
generation of curly-haired members of Population B, preventing them from
mating and contributing to the next generation. The sad but affectionate remaining
members of the population breed randomly. What would be the new f(curly
haired) following this second generation of loss?
Punnett square for HardyWeinberg
Females
A (p)

B (q)
2

Males

A (p) AA (p )

AB (pq)

B (q) AB (qp)

BB (q2)

Q5
You are recording from neuron A. It has the voltage-dependent channels needed to fire an
action potential. Neuron A is receiving a number of inputs:

EK= -80mV; ECl = -60mV; ENa= +60mV; Vrest = -60mV; Total input conductance at rest
is 14nS; Threshold for action potential firing is -50mV; Assume that gCl at rest is very
low.
a. What are the values of gNa and gK at rest?
b. Suppose that Neuron B evokes an EPSP with a reversal potential of -10mV and a
synaptic conductance of 2nS. What is the membrane potential at the peak of the
EPSP?
c. Will Neuron A fire an action potential in response to the Neuron B EPSP?
d. Calculate the membrane potential at the peak of the EPSP if the reversal potential
were +60mV. Will the cell fire?
e. Neuron C evokes an EPSP with a reversal potential of -10mV and a synaptic
conductance of 12nS. Will Neuron A fire in response to input from neuron C?
f. Cell D produces an increase in gCl of 12nS. What does this do to the membrane
potential of A?
g. Cell E produces an increase in gCl of 120nS. What does this do to the membrane
potential of A?
h. What happens if we activate cell D along with cell C? Will A fire?
i. What happens if we activate cell E along with cell C? Will A fire?
j. Explain in words the implications of the above.

Q6
[Calculator required] An antibody-based diagnostic kit can operate in three different
modes, shown to have the following false-positive (FP) and false-negative (FN) rates:
Mode
A
B
C

FP
0.005
0.15
0.01

FN
0.5
0.15
0.5

You plan to use these kits to screen for infections in a certain population. You already
know that Infection-1 has an incidence of 5% and Infection-II has an incidence of 25%,
and that the two are not correlated. You can only test each individual once for each
infection.
Which mode of testing would be most informative when testing for: Infection-I?
Infection-II?
This is a quantitative question, show all your calculations. The table of logarithms in base
2 might be useful.

log2(x)
x

inf.

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

-6.64 -5.64 -5.06 -4.64 -4.32 -4.06 -3.84 -3.64 -3.47

0.1 -3.32 -3.18 -3.06 -2.94 -2.84 -2.74 -2.64 -2.56 -2.47 -2.4
0.2 -2.32 -2.25 -2.18 -2.12 -2.06 -2
0.3 -1.74 -1.69 -1.64 -1.6

-1.94 -1.89 -1.84 -1.79

-1.56 -1.51 -1.47 -1.43 -1.4

-1.36

0.4 -1.32 -1.29 -1.25 -1.22 -1.18 -1.15 -1.12 -1.09 -1.06 -1.03
0.5 -1

-0.97 -0.94 -0.92 -0.89 -0.86 -0.84 -0.81 -0.79 -0.76

0.6 -0.74 -0.71 -0.69 -0.67 -0.64 -0.62 -0.6

-0.58 -0.56 -0.54

0.7 -0.51 -0.49 -0.47 -0.45 -0.43 -0.42 -0.4

-0.38 -0.36 -0.34

0.8 -0.32 -0.3

-0.29 -0.27 -0.25 -0.23 -0.22 -0.2

0.9 -0.15 -0.14 -0.12 -0.1

-0.18 -0.17

-0.09 -0.07 -0.06 -0.04 -0.03 -0.01

Q7
It is well established that a hormone H when applied to cells causes them to increase in
size without undergoing cell division. Recently a new molecule X has been described
that is generated within cells and whose effect is cells increasing in size.
Suggest two approaches to establish the relationship if any, between H and X in the
regulation of cell size.

Q8
The asymmetric distribution of proteins or
small molecules among different sub-cellular
membrane bound compartments is frequently
seen in eukaryotic cells. A small molecule S is
almost entirely localized to compartment B. S
is synthesized by enzyme E1 and degraded by
enzyme E2.
Describe cellular mechanisms by which the
distribution of S can be restricted to
compartment B.
Q9
a. A hypothetical transcription factor TF420 binds the double stranded DNA sequence
GCGCGC with a Kd of 4.2 nM. Would you anticipate TF420 to bind a double stranded
RNA containing the sequence GCGCGC? Design an experiment that will test whether it
binds or does not bind a GCGCGC-containing dsRNA sequence. Explain what results
you would see depending on whether it does or does not bind.
b. Another protein RBP101 is known to bind double-stranded RNA. The protein seems to
be selective in its binding, having a strong affinity to only a few dsRNA sequences out of
a large number tested. However, researchers are unable to find a consensus sequence
among these strong binders. Provide a hypothesis for how, on the one hand, binding can
be selective whereas, on the other hand, there is no consensus sequence. Design an
experiment to test this hypothesis.
Q10
Examine the following two assertions and design an experiment, in each case, that
provides proof for the assertion:
a.
b.

A regenerating salamander limb always does so in the distal direction and there is
no attempt to replace missing parts.
The Hydra head-region produces a substance inhibitory for regeneration that falls
off with distance from the head.

Your answers should be written out and please NOT use figures!

Q11
There are many possible definitions of a protein domain. Give a reasonable scientific
definition in each of the following categories, and also a test (algorithm or experiment) to
determine the domain breakup of the protein according to each definition:
a. Structural

b.

Functional

c.

Folding

d.

Evolutionary

Q12
The relatively recent paper by Reich et al., 2010 detailed genetic data and analyses for a
fossil tooth found in Siberia. This paper and the data and analyses presented therein
suggested a new species of archaic humans, called the Denisovans. This and other
studies on the Denisovan genome suggest the relationships shown in panel A (adapted
from Krause et al., 2010).
A

a.

Given the figure in panel A, draw a proposed model for human evolution (with
timings) on the upper map on the next page, detailing fossils (or rough
distributions) of human ancestors from Australopithecus species to Homo species.
Also include the Neanderthals and the Denisovans.

b.

Do the results in panel B support a multi-regional model for human evolution? Do

you have enough information to draw this inference? If not, what additional
information might you need to support the multi-regional model?

c.

What undiscovered (hypothetical) fossils would provide support for the multiregional hypothesis, given the context of the Denisovans and the Neanderthals?
Include several fossils (on the lower map on the next page) of different ages.
Assume you are able to extract genetic data from these fossils. Now place these
fossils into a new version of the phylogenetic tree above, such that the tree
supports the multi-regional hypothesis. Re-draw the tree to support the Out of
African Hypothesis.

d.

Take an educated guess about how much Neanderthal ancestry your genome may
possess.