Thesis_XP_final

Fitness study on mutation-accumulation lines of
Saccharomyces cerevisiae
Name: Xiangyu Peng
Mentor: Mark L. Siegal

A thesis in fulfillment of the Masters in Biology Degree
Summary
It was generally believed the major effects of spontaneous mutation are deleterious.
But recent studies in many species have found unexpectedly high frequencies of
beneficial mutations. To better understand the effects of mutations on fitness at
different phases of the life cycle in budding yeast, Saccharomyces cerevisiae, we used
2000-generation mutation-accumulation (MA) lines. These MA lines derived
independently from a single diploid ancestor. From each of 31 MA lines and the
ancestral line, a single haploid progeny of mating-type a was obtained. We converted
these 32 lines into haploid lines of mating-type α, as well as homozygous a/α diploid
lines. On the 96 lines we performed micro-colony growth-rate assays to determine
their relative fitnesses. The results showed evidence of beneficial mutations, in that
some MA lines grew faster than the ancestral line of the same ploidy and mating type.
Surprisingly, the mean fitness of haploid α is higher than haploid a, whereas they
were expected to be at the same level. We also found some false positive diploids,
which were diploid determined by the halo mating type test, but actually they are

haploids.
Introduction
Spontaneous Mutation
Mutation, which is the source of genetic variation, conditions the response to
selection and adaptation through natural selection. Most researchers consider
mutations are usually deleterious. In 1930, Dr. R. A. Fisher proposed this theory,
introducing a geometric model that implied the deleterious effects of most mutations
(Fisher, 1930). Moreover, some molecular data on the ratio of nonsynonymous to
synonymous substitution rates pointed out that most nonsynonymous mutations have
deleterious effects on fitness (Eyre-Walker et al., 2002). Several studies of mutation
accumulation also detected a decline in mean fitness, indicating that most
accumulated mutations are deleterious. However, some recent studies suggest that
spontaneous mutations might have beneficial effects more commonly than we thought
before. A mutation-accumulation study on Arabidopsis thaliana found that 50% of
spontaneous mutations were beneficial (Shaw et al., 2002). An analysis of
1012-generation mutation-accumulation (MA) lines of yeast found that 5.75% of the
fitness-altering mutations were beneficial (Joseph & Hall, 2004). Spontaneous
antibiotic mutations in Pseudomonas fluorescens also showed ~2%-3% of non-neutral
mutations were beneficial when lacking antibiotics (Kassen & Bataillon, 2006).

Although different estimations of beneficial rate and distribution of beneficial
mutations have been performed in many organisms, just few MA experiments
identified beneficial mutations. Such different estimates even derive from studies of
the same species in similar environments (Zeyl and Devisser, 2001; Joseph and Hall,
2004; Dickingson, 2008; Hall et al. 2008). This might be due to different evolutionary
potential of the initial genotypes that were used in these experiments. Now the extent
to which spontaneous mutations are deleterious or not remains uncertain. The relative
effect sizes of deleterious and beneficial mutations are also generally unknown.
Mutation-Accumulation Lines
The mutation accumulation approach lets mutations randomly accumulate under
benign conditions in a series of sublines derived from an inbred base population
(ideally a single completely homozygous individual) (Bataillon, 2000). We acquired
diploid MA lines from Dr. David W. Hall at University of Georgia. Dr. Hall
established the MA lines from an ancestral strain, which was produced by sporulating
a diploid strain (DBY4974/DBY4975), to yield a haploid strain of genotype ade2,
lys2-801, his3-Δ200, leu2-3.112, ura3-52, Gal+, ho. Then the haploid was
transformed with HO plasmid to induce diploidization. The resulting diploid strain
became the ancestral strain for mutation accumulation (Joseph and Hall, 2004). The
ancestral strain was homozygous at all loci except the mating-type locus, which was
a/α. And it is mutant at only a few loci, none of which are suspected to affect DNA

integrity or repair, so it is a good representative of mutational process in a wild-type
strain. In addition, the ade2 mutation causes the accumulation of a metabolite in the
adenine biosynthetic pathway. In the presence of oxidative respiration, the colonies
turn reddish and in its absence, colonies are white. So it helps to circumvent petite
mutations, which have been a problem in previous yeast MA experiments, visually by
only transferring red colonies. The mitochondrial mutants, or petites, are defective in
the respiratory chain, and therefore exhibit a substantial reduction in fitness and are
unable to grow on media containing only non-fermentable carbon sources (such as
glycerol or ethanol) (Bernardi, 1979).
One hundred and fifty-one MA lines were established from the ancestor. Each MA
line was established by single-cell passage on YPD solid medium (1% yeast extract, 2%
peptone, 2% dextrose and 2% agar). One passage is conducted by randomly selecting
one nonwhite colony then streaking it onto a new YPD plate. Each plate was
incubated at 30°C for 48 hours until the next passage. 100 passages were performed
and the resulting strains are 2000-generation descendants of the ancestral strain.
These 2000-generation descendants with the ancestral strain are the diploid MA lines
we acquired from Dr. Hall.
High-Throughput Microcolony Growth Assay
A novel method was developed in the Siegal lab to measure microcolony growth by

time-lapse bright-field microscopy (Levy et al., 2012). It overcomes some limitations
in microbial fitness assays. Exponentially growing cells are plated at a low density in
rich, liquid medium on glass-bottomed micro-well plates and allowed to grow into
isolated microcolonies of up to ~100 cells. During this growth period, 1-hour
time-lapse images of ~3,000 low-magnification fields are captured in parallel
allowing for simultaneous observation of ~105
microcolonies. Custom-written image
analysis software tracks changes in area over time, and these measurements are used
to calculate the specific growth rate of each microcolony (the change in the log of the
area per hour) (Figure 1) (Levy et al., 2012).
Figure 1. High-Throughput Microcolony Growth Assay to measure growth rate
in yeast. (A) A schematic of the assay. Cells in logarithmic growth are plated at a low
density on a glass-bottomed multi-well plate. Low-magnification time-lapse
0 h
5 h
10 h
A B

bright-field images are captured in a highly parallelized manner. (B) Isogenic cells
grow at different rates. The software tracks every growing single cell. On the left is
shown a portion of one field of time-lapse images. On the right is shown the output of
image analysis.
We used the MA lines and the new growth-rate assay to study the effects of
spontaneous mutations on fitness in cells of different ploidy and mating type in
budding yeast. To accomplish this goal, we transformed the 32 haploid a strains with
a plasmid encoding an inducible source of HO endonuclease, which is required for
mating-type switching. Induction of HO expression may produce colonies of all three
types (haploid a, haploid α and diploid). In cases when no diploid was obtained,
mating of the two haploids was used to generate a diploid. We performed halo tests to
determine the mating types of the colonies, and selected one representative haploid
α and one representative diploid for each original MA strain to establish a 96-well
master plate for micro-colony growth-rate assays.
Materials and Methods
Transformation
Prior work in Siegal lab had produced haploid a derivatives by sporulation of the
diploid MA lines. So we first focused on making the haploid α and diploid strains by
performing mating-type switching using an inducible pGal-HO plasmid. We used the

PLATE transformation protocol, where ‘PLATE’ is an acronym of the names of
ingredients in the transformation solution: PEG, Lithium Acetate, Tris and EDTA.
Each MA strain (haploid a) was saturated at 30°C in YPD medium (1% yeast extract,
2% peptone and 2% dextrose). After saturation, cells were collected by centrifuging
0.5 ml of the saturated culture. The pelleted cells were mixed with 10 µl of sonicated
salmon sperm DNA (10 mg/ml stock), 7 µl of the pGal-HO plasmid DNA (145 ng/µl)
and 500 µl of PLATE solution (40%PEG3350 (w/v), 100mM lithium acetate, 10mM
Tris, pH=7.5, 0.4mM EDTA). The mixture was incubated at RT for 15 min, then heat
shocked at 42°C for 15 min, and then immediately cooled on ice for 2 min. The
transformed cells were pelleted and resuspended in 200 µl of sterile water by
pipetting gently and thoroughly and spread onto a SC-URA plate (0.67 % yeast
nitrogen base without amino acids, 0.2 % Uracil knockout drop-out mix, 2% dextrose
and 2 % agar), which selected against ura3. Only the cells that were transformed with
HO plasmid would survive on the SC-URA plate. The plate was incubated at 30°C
until colonies appeared. One red colony was randomly selected and spread onto a new
SC-URA plate. The resulting colonies formed on the new SC-URA plate were used
for the switching assay. 32 strains were successfully transformed.
Switching Assay
There are two solutions involved switching assay. Solution 1 is Minimal media (0.67%
yeast nitrogen base without amino acids and 0.2 % Uracil knockout drop-out mix),

0.1% glucose and 2% raffinose. Solution 2 is Minimal media, 0.1% glucose and 2%
galactose. Raffinose will derepress the GAL promoter and galactose is used to induce
transcription of sequences fused to the GAL10 promoter, which will induce the
switching of mating type. Each transformed strain containing the HO plasmid was
grown in 5 ml Solution 1 overnight at 30°C. The next morning, the overnight culture
was resuspended and 1 ml of culture was added to 4 ml of Solution 2. Then the
mixture was grown at 30°C for 5 hours to induce mating type switching for
approximately one generation. After 2, 3, 4 and 5 hours of growth in the presence of
galactose, a 1-ml aliquot of cells was taken out and placed on ice. At the end, all
aliquots were combined. Based on the ODλ=600, 104
times dilution in YPD is optimal.
200 µl of the dilution was spread onto YPD plate and incubated at 30°C until colonies
appeared.
After the switching assay, some cells could still contain the pGal-HO plasmid. The
functional HO gene will make the cells switch mating type remarkably efficiently,
within a few cell divisions after the spore germinates (Klar, 2010). In order to make
the cells stable in their mating type, we streaked the desired colonies onto 5-FOA
plates (0.67 % yeast nitrogen base without amino acids, 0.2 % Uracil knockout
drop-out mix, 0.005% Uracil, 2% Dextrose, 0.1% 5-FOA and 2% agar) to select
against URA+ after the halo test.

Halo Mating Type Assay
We used this method to determine the mating type of each single-cell colony. There
are two tester strains: DBY7730 (aka RC634a) MATa and DBY7442 (aka XT1-20A)
MATα. These two strains all contain sst mutations, which make them super-sensitive
to pheromone. When exposed to pheromone of the opposite mating type, the cells
arrest. The two tester strains were grown overnight in YPD. The next morning the
overnight culture was diluted 10 times with sterile water. Then dilutions were spread
on YPD plates, one for each tester strain, and incubated at 30°C for 30 min to 45 min.
When the tester plates were ready, the switching plate, which contains unknown
mating type colonies, was replicated on a new YPD plate first then on the two tester
plates by replica-plating. All three plates were incubated at 30°C overnight. The
replicate switching plate was used for the following experiments.
A haploid α colony shows a halo space around it on RC634a tester plate whereas a
diploid colonyi shows no halo space on either of the tester plates. After the halo test,
haploid α and diploid were streaked onto 5-FOA plates to select against URA+. We
also used SC-URA plates as control plates to test the loss of pGal-HO plasmid.
Yeast Mating
Due to the surprisingly low yield of diploids (only 8 of 32 strains obtained diploids
from the switching assay), we also performed yeast matings to obtain the diploids of

the rest of the strains. First, haploid a and haploid α were grown overnight in YPD at
30°C. Then 200 µl of each haploid culture was mixed and cells were pelleted by
centrifuging. The excess media was poured out gently. By using the remaining media,
the cells were resuspended and 15 µl of the suspension was spotted on a YPD plate.
The plate was incubated for 5 hours and then the whole spot was scraped off and
resuspended in 1 ml of sterile water. To yield an optimal amount of colonies on a
YPD plate, the suspension was diluted 105
times and 200 µl of the dilution was spread
on a YPD plate. This plate was grown at 30°C for two days. Diploids were identified
by halo test, after which HO plasmid was removed by selecting against URA+ on
5-FOA plate.
Sporulation Assay
When in the presence of a poor carbon source environment, diploid cells of the yeast
S. cerevisiae will undergo meiosis and package the haploid nuclei produced in
meiosis into spores. So we used the sporulation protocol to identify whether a strain
acquired from the yeast mating assay was truly diploid. However, sporulation is just a
sufficient, but not necessary condition for identifying diploids. A strain without the
ability of sporulation does not mean that it is not a diploid, especially in MA strains,
since some accumulated spontaneous mutations might harm the cells’ ability to
sporulate. The sporulation assay served as a preliminary test for the candidate
diploids.

First, the candidate diploid was grown overnight at 30°C in YPD medium. The next
morning, 60 µl of the culture was added to 3 ml of YPD medium in a new tube and
was shaken at 30°C for 6 hours. After the incubation, the cells were pelleted by
centrifuging and the supernatant was removed. Then 2.5 ml of sporulation media was
added to the pelleted cells and well mixed. The cells were starved for 5 days at room
temperature on a shaker. Microscopic examination was used to detect the spores after
the sporulation assay.
Fluorescence-activated cell sorting (FACS) with Sytox Green
Neither the halo test nor the sporulation test can give the most convincing conclusion
of whether or not a candidate is truly diploid. FACS was performed for candidates to
determine the DNA content of the test strains, which would give clear results of their
ploidy.
An overnight culture of the candidate diploid was obtained by incubating at 30°C in
YPD. 5*106
cells were pelleted from the culture and washed one time with 400 µl of
ddH2O. Following the wash, the cells were resuspended in 400 µl of ddH2O and
transferred to a 1.5 ml Eppendorf tube. The suspension was then sonicated with small
probe, mixed with 950 µl of 100% EtOH and incubated for 1 hour at RT. After fixing
with EtOH, cells were pelleted and washed with 800 µl of 50 mM Na citrate (1.47%

sodium citrate dehydrate, pH 7.2 and filtered with 0.45 µm filter). The washed pellet
was resuspended in 0.5 ml Na citrate containing 0.25 mg/ml RNase A and incubated
at 50°C overnight. Next day, 10 µl of proteinase K (100mg/ml) was added to the
solution and the sample was returned to 50°C for 2 hours. Then it was sonicated and
mixed with 0.5 ml Na citrate containing 4 µM Sytox Green. The mixture was
transferred to a Falcon 2054 tube and incubated in the dark for 1 hour. When the
incubation was done, FACS analysis was performed by the Becton Dickinson
FACSAria cell sorter.
Microscopy
After isolation of the 32 strains each of haploid α and diploid, we made one stock
plate and 10 master plates plate containing 32 strains of a mating types in the first
four columns, α mating types in the middle four columns and diploids in the last four
columns. The stock plate was a 96-well plate with each well containing 200 µl
mixture of 100 µl of the corresponding saturated cell culture and 100 µl of 30%
glycerol. The master plate was a 96-well plate with each well containing 20 µl of the
same mixture as the stock plate. Before starting the growth rate assay, we added 180
µl of SC+adenine media (0.67 % yeast nitrogen base without amino acids, 0.2 %
Uracil knockout drop-out mix, 0.005% Uracil, 0.006% Adenine and 2% Dextrose) to
each well and let cells grow at 30°C in a shaking incubator for 24 hours. The next day
we transferred 10 µl of the culture from each well to a new 96-well plate and added

190 µl of SC+adenine media to each well. The cells in new plate were grown at 30°C
in a shaking incubator for 48 hours.
Then the cells were sonicated and diluted to a final concentration of 2*104
cells/ml.
Glass-bottomed 384-well plates were coated with 50 µl of 200 mg/ml Concanavalin A
for 90 min. Wells were washed twice with 200 µl of sterile water and 75 µl of cells
were plated per well. In order to control for any well effect, a Tecan Freedom EVO
liquid handling robot was used to randomize the well locations of the samples. There
were four replicates of each sample per plate. Plates were sealed tightly with an
optically clear film and spun at 1500 RPM for 2 min. Before placing the samples on
the microscope, the top and bottom surface of the 384-well plate were wiped with
kim-wipe and dusted using compressed air to remove any particles that may interfere
with the Nikon Perfect Focus System (PFS). The Nikon microscope equipped with
PFS captured micrographs from four fields of each well every hour. For each sample,
there were sixteen fields in total from four different wells. ~200-500 microcolonies
were detected in each field (Figure 2).

Figure 2. Histogram of frequency of the number of microcolonies in each well. A
large portion of wells has ~200-500 microcolonies, which was good for acquiring
average growth rate of each sample.
Results
The microcolony growth assay was performed twice and the results were combined to
obtain the overall estimation. Statistical analyses were performed using R
programming language and plots were made using the package ‘ggplot2’. The plate
and well (and their interaction) conditional means were estimated by residual

maximum likelihood (REML) and subtracted from the original mean of each well of
the two plates to control for plate and well effects. The mean growth rates of the three
cell types of each of the 32 strains are shown in Figure 3. We observed an increased
mean growth rate of diploid and haploid α relative to the original haploid a MA lines.
The higher growth rate in diploids was expected, but the higher growth rate of haploid
α relative to haploid a was unexpected. To our knowledge, a growth advantage of α
cells has never been reported in the literature. It might be that the
pheromone-responsive mating pathway, which is known to exert a cost on haploid
cells (Lang et al., 2009), has a lower cost in α cells than in a cells.
The growth rate of haploid a and haploid α MA lines was on average lower than the
haploid a and haploid α versions of the ancestral strain, respectively, although some
lines showed increased growth relative to the ancestor. By contrast, the growth rate of
diploid MA lines was on average higher than the diploid ancestor. The decrease of
mean growth rate of haploid a and haploid α is in accordance with the theory of
deleterious mutations. But diploidy might buffer the deleterious effect of spontaneous
mutations, and even make them beneficial.

Figure 3. Plot of mean growth rate of three cell types of 32 strains. The same cell
type of different strains is in the same x-axis coordinate position. Y-axis is the
normalized mean growth rate. The same color represents the same strain and three
cell types of the same strain were connected by lines to show the difference of mean
growth rates among the different cell types of the 32 strains. Ancestral strains are
highlighted in black.
Variance of growth rate is also a parameter that can affect fitness (Levy et al., 2012).

Figure 4 shows the pooled standard deviation plotted versus the mean of each strain.
For each cell type and MA line combination, the pooled standard deviation was
calculated based on the standard deviations from two plates. The scatter plot indicates
that the variance of growth rate depends on mean. The variance of growth rate is
higher for haploid a MA lines than for the other two cell types, even when taking into
account the dependence of variance on mean, as is done in Figure 4 by lowess
regression. The residual distance of each point from the regression is calculated.
These residuals of standard deviation represent a measure if variance controlled for
the mean (Levy et al., 2008). For each cell type, we made a rank plot for 32 strains
(Figure 5) to better visualize the change of variance and observed that the variance of
haploid a and haploid α MA lines was higher than the haploid a and haploid α
versions of the ancestral strain, respectively. In contrast, the variance of diploid MA
lines was lower than the diploid ancestral strain. In combined with mean growth rate
results, it is very likely that the random mutations tend to decrease the growth rate and
make growth more variable in haploid a and haploid α cells. Instead, diplody could
buffer the deleterious effect and make the growth faster and less variable.

Figure 4. Scatter plot of mean and standard deviation for each cell types.
Different cell types are distinguished by different colors. Red dots represent diploid,
green dots represent haploid a and blue dots represent haploid α. Ancestor haploid a
and haploid α are highlighted in black and diploid is highlighted in purple.

Figure 5. Rank plot of residuals for each cell type of 32 strains. Different strains
are distinguished by different colors and ancestor is highlighted in black. X-axis is
different strains and Y-axis is the residuals, which represent a measure of variance
controlled for the mean.
Because we suspected that some supposed diploids were acting like haploids, we first
reviewed all the micrographs of haploid α and diploid. Due to the different budding
patterns of haploid and diploid (Figure 6), we could easily visualize whether
suspected non-diploids had abnormal growth patterns. Micrographs showed that there
the budding pattern in haploid α cells appeared normal, but 12 of 32 diploid strains
budded like haploids do. We therefore performed halo tests again to determine

whether the initial assignment of the strains as diploids was due to human error in the
halo test or perhaps instead to some mutation that either inactivated pheromone
production in a haploid or changed the budding pattern of a diploid.
Figure 6. Two distinct budding patterns. Haploid cells bud in an axial pattern, in
which each bud emerges adjacent to the site of the previous bud. Diploid cells bud in
a
bipolar pattern, in which successive buds can form at opposite poles of the cell.
The halo test showed that 8 of 12 supposed diploid strains were actually haploid α,
which was probably caused by mistakes in making the master plates or by
misjudgement of the original halo assays due to unclear halos around the colonies.
However, there were 4 of them that showed no halo (Figure 7), which indicated that
they should be diploid strains. We considered these four strains — MAD.0, MAD.11,
MAD.15 and MAD.119 — as candidate diploid strains.
We extracted the mean growth rates of three cell types of these four strains (Figure 8).

It is obvious that the haploid α and diploid of these strains had almost the same level
of mean growth rate.
Figure 7. Halo test for haploid strains. No halo space was around the labeled strains,
which indicates that these four strains were diploid. The names on top of the figure
indicated the tester strain spread on each plate.

Figure 8. Plot of mean growth rate of four candidate strains. Haploid α of the four
strains has almost the same level of mean growth rate relative to diploid.
To further determine the actual cell type of these four candidate diploid strains, we
performed sporulation assays. If these four candidates can sporulate, they should be
actual diploid strains. These candidates showed no sign of sporulation after shaking
five days in sporulation solution. However, no sporulation does not necessarily mean
that they are not diploid strains, because the sporulation function might be
compromised by some spontaneous mutation.

The most convincing way to determine the ploidy of a cell is DNA content test. So we
performed FACS with Cytox Green (Figure 9). A haploid strain will show a major
portion of peak in 1N phase and a small portion of peak in 2N phase. A diploid strain
will show a major portion of peak in 2N phase and a small portion of peak in 4N
phase. The four candidates are all haploid strains, which means that the halo tests for
these strains were incorrect. This might be due to a defect in emitting pheromone in
cells caused by spontaneous mutations.
Figure 9. FACS Analysis with Sytox Green for the four candidate diploid strains.
All four candidates are haploid strains
To find out whether haploid α has a growth advantage over haploid a, we performed

the microcolony growth assay for three pairs of isogenic strains of the two mating
types: BY4741 (haploid a), BY4742 (haploid α), FY4 (haploid a), FY5 (haploid α),
W303a (haploid a) and W303α (haploid α). Each strain had 16 replicate wells in a
96-well plate instead of a 384-well plate. There was no consistent pattern from the
experiment (Figure 10).
Figure 10. Plot of mean growth rate of the three pairs of test strains. There was a

significant higher mean growth rate between BY4741 and BY4742. But the rest two
pairs had almost the same level of mean growth rate.
Discussion
In this study, we performed transformation, switching assays, mating assays, halo
assays and micro-colony growth rate assays starting with 32 original haploid a MA
strains. We observed a decrease of mean growth rate and an increase variance in
growth rate of haploid a and haploid α MA lines relative to original haploid a and
haploid α ancestor. This is consistent with the deleterious theory of spontaneous
mutations. In contrast, diploidy might be able to buffer the deleterious effect of
spontaneous mutations due to the increase of mean growth rate and the decrease
variance in growth rate of diploid MA lines relative to original diploid ancestor.
We also found 4 false-positive diploid strains from the halo assays and a previously
unreported growth advantage of haploid α cells over otherwise genetically identical
haploid a cells. In order to avoid false-positive diploid strains, we should adopt a
more direct method to obtain diploids. For example, after the yeast mating assay, we
could pick out zygotes under microscopy by using the micromanipulator and grow
them on YPD plates.
We are also interested in why the mean growth rate of haploid α strains was higher

than that of haploid a strains. Haploids of the same strain are thought to grow at the
same rate, because they have the same genetic background except at the MAT locus.
The haploid α version of the ancestor has a higher mean growth rate than the haploid
a version, suggesting that it is not a bias in the effects of accumulated mutations
driving the disadvantage of a cells. Instead, we propose that the mating pathway
might have different costs in α cells and in a cells, and that the costs are possibly
dependent on the genetic background. To test this idea, we have acquired 20 new
pairs of isogenic haploid a and haploid α strains on which to perform micro-colony
growth rate assays.
References
Bataillon, T. (2000). Estimation of spontaneous genome-wide mutation rate
parameters: Whither beneficial mutations? Heredity 84: 497–501.
Bernardi, G. (1979). The petite mutation in yeast. Trends Biochem. Sci. 4: 197–201.
Chant J., Mischke M., Mitchell E., Herskowitz I., Pringle J.R. (1995). Role of Bud3p
in producing the axial budding pattern of yeast. J Cell Biol. 129(3):767-78.
Drake, J.W. (1991). A constant rate of spontaneous mutation in DNA-based microbes.
Proc. Natl. Acad. Sci. USA 88: 7160–7164.
Drake, J. W., B. Charlesworth, D. Charlesworth and J. F. Crow (1998). Rates of
spontaneous mutation. Genetics 148: 1667–1686.
Egel, R., Willer, M., Kjaerulff, S., Davey, J., Nielsen, O. (1994). Assessment of

pheromone production and response in fission yeast by a halo test of induced
sporulation. Yeast. 10(10):1347-54.
Fisher, R. A. (1930). The Genetical Theory of Natural Selection. Oxford University
Press, Oxford.
Garcia-Dorado, A., C. Lopez-Fanjul and A. Caballero. (1999). Properties of
spontaneous mutations affecting quantitative traits. Genet. Res. 74: 341–350.
Gu, Z. L., L. M. Steinmetz, X. Gu, C. Scharfe, R. W. Davis et al. (2003). Role of
duplicate genes in genetic robustness against null mutations. Nature 421: 63–66.
Hall, D.W., Joseph, S.B. (2010). A high frequency of beneficial mutations across
multiple fitness components in Saccharomyces cerevisiae. Genetics. 185(4):1397-409.
Hall, D.W., Mahmoudizad, R., Hurd, A.W., Joseph, S.B. (2008). Spontaneous
mutations in diploid Saccharomyces cerevisiae: another thousand cell generations.
Genet Res (Camb). 90(3):229-41.
Joseph, S.B., Hall, D.W. (2004). Spontaneous mutations in diploid Saccharomyces
cerevisiae: more beneficial than expected. Genetics. 168(4):1817-25.
Kawai, S., Hashimoto, W., Murata, K. (2010). Transformation of Saccharomyces
cerevisiae and other fungi: methods and possible underlying mechanism. Bioeng Bugs.
1(6):395-403.
Klar, A. J. (2010). The yeast mating-type switching mechanism: a memoir. Genetics.
186(2):443-9.
Korona, R. (1999). Genetic load of the yeast Saccharomyces cerevisiae under diverse

environmental conditions. Evolution 53: 1966–1971.
Kussell, E., Leibler, S. (2005). Phenotypic diversity, population growth, and
information in fluctuating environments. Science 309: 2075–2078.
Lang, G.I., Murray, A.W., Botstein, D. (2009). The cost of gene expression underlies
a fitness trade-off in yeast. PNAS. 106(14):5755-5760.
Levy, S. F., Siegal, M.L. (2008). Network hubs buffer environmental variation in
Saccharomyces cerevisiae. PLoS Biol. 6(11):e264.
Levy, S. F., Ziv, N., Siegal, M.L. (2012). Bet hedging in yeast by heterogeneous,
age-correlated expression of a stress protectant. PLoS Biol. 10(5):e1001325.
Zeyl, C., and J.DeVisser. (2001). Estimates of the rate and distribution of fitness
effects of spontaneous mutation in Saccharomyces cerevisiae. Genetics 157: 53–61.

Thesis_XP_final

More Related Content

What's hot (19)

Similar to Thesis_XP_final (20)

Thesis_XP_final