Unveiling nonessential gene deletions that confer significant morphological phenotypes beyond natural yeast strains
© Yang et al.; licensee BioMed Central Ltd. 2014
Received: 21 July 2014
Accepted: 15 October 2014
Published: 25 October 2014
Phenotypes are variable within species, with high phenotypic variation in the fitness and cell morphology of natural yeast strains due to genetic variation. A gene deletion collection of yeast laboratory strains also contains phenotypic variations, demonstrating the involvement of each gene and its specific function. However, to date, no study has compared the phenotypic variations between natural strains and gene deletion mutants in yeast.
The morphological variance was compared between 110 most distinct gene deletion strains and 36 typical natural yeast strains using a generalized linear model. The gene deletion strains had higher morphological variance than the natural strains. Thirty-six gene deletion mutants conferred significant morphological changes beyond that of the natural strains, revealing the importance of the genes with high genetic interaction and specific cellular functions for species conservation.
Based on the morphological analysis, we discovered gene deletion mutants whose morphologies were not seen in nature. Our multivariate approach to the morphological diversity provided a new insight into the evolution and species conservation of yeast.
Evolution has produced remarkably complex and diverse living organisms with different morphological phenotypes observed in shape, size, and other traits. This morphological variation is important for their survival during environmental disruption, and many biologists aim to clarify how organisms evolved their phenotypic variation throughout their long evolutionary history.
The budding yeast Saccharomyces cerevisiae is a leading model organism used in genetics and systems biology because it has cellular processes in common with many eukaryotic cells. After the whole genome of the laboratory yeast strain was sequenced , most yeast genes were functionally annotated, providing insights into the relationship between genotype and phenotype. High genetic variance was found in different yeast subgroups based on analyses of yeast strains isolated from different ecological niches [2, 3]. Natural yeast strains also exhibited high phenotypic variation based on the analysis of yeasts cultured under various environmental conditions . These results described the relationship between phenotype and genetic background, which provided insights into the origins of natural phenotypic variation. However, the relationship between the phenotype and genotype in natural strains remains unclear, as how an individual gene influences the phenotypic variation within the species is unknown. Furthermore, most previous research was restricted to fitness  and gene expression  phenotypes.
In this study, a high-dimensional image-processing program CalMorph was used on digital microscopic images to ascertain cell shape, actin, and nuclear DNA morphology. We analyzed the morphological variations with a generalized linear model (GLM), an extension of the normal linear model , by incorporating various probability distribution models. The statistical models were set to assess the effects of a homozygous gene deletion of yeast diploid on cell morphology.
Mosaic segregants are more variable than their pure parental strains
To evaluate the morphological diversity in yeast, we focused on two aspects: the standard deviation in each parameter and the population expansion in the orthogonal phenotypic space that was degenerated in dimension from high-dimensional morphological traits. The diverse population could be expanded in orthogonal phenotypic space.
To ascertain the degree of global morphological variance, we performed principal components analysis (PCA), a statistical procedure that uses an orthogonal transformation. An advantage of comparing in the degenerated orthogonal space is that one can exclude bias caused by the correlation between the morphological parameters. We found that the morphological distribution of the mosaic segregants was broader than that of the parental strains in the principal component (PC)1 and PC2 spaces (Figure 2D). The mosaic segregants also showed a broader distribution in the spaces between any other pairs from PC1 to PC4 (Additional file 1: Figure S1). Thus, our analyses confirmed that the mosaic strains contained a higher morphological variance than that of the parental strains and validated our procedure.
Homozygous gene deletion strains have higher morphological variance than natural strains
Identification of mutants with a higher number of morphological changes than natural strains
Properties of the 36 heteroclite mutants
Although the one sample test using the natural strains identified 36 heteroclite mutants out of 110 mutants, the average Mahalanobis distance from the parental strains was not significantly different (Mann–Whitney U-test, P <0.05; Figure 4B), suggesting that the difference between these groups was not due to the extent of morphological difference but rather to different morphological profiles. Because each PC represented different cell morphological features, we looked for differences in the PCs. Then we found that these 36 genes were distributed differently from those in the PC1 and PC2 spaces (red dots in Figure 4C). Some of the deletion mutants with larger (e.g., ctf4) or smaller (e.g., cog1) cell sizes (higher or lower PC1 scores, respectively) or a higher accumulation of cells with delocalized actin patches (e.g., ypk2; lower PC2 scores) were defined as the heteroclite mutants, which indicated that a loss-of-function of the 36 corresponding genes (hereafter we refer them as “heteroclite genes”) resulted in morphological changes in specific traits.
Fitness analysis of the 36 heteroclite mutants
Yeast mutants with an unusual morphology tend to suffer from growth defects. Deletion mutants with an abnormal morphology, not present in the natural yeast strains, may be missing from the natural population due to their slow growth. We examined this possibility in the 36 heteroclite mutants and in the remaining 74 mutants. Fitness was assessed for the nonessential gene deletion strains in rich media  to compare the degree of fitness. We found that number of deletion mutants with significantly decreased fitness was similar; 22 (60%) and 39 (58%) gene deletion mutants conferred decreased fitness compared to the wild-type strains in the 36 heteroclite and 74 mutants, respectively. An alignment of these mutants according to their degree of fitness is shown in Additional file 9: Figure S7A. The distribution of the deletion mutants (Additional file 9: Figure S7B) indicated that the fitness of most of the 4718 strains was distributed around a central fitness of 1, whereas the heteroclite and the rest of the gene deletion mutants were distributed at a lower fitness level. No significant differences were observed between the heteroclite mutants and the other mutants (Mann–Whitney U-test, P =0.72). Thus, many morphological mutants had decreased fitness, but this was not more common in the 36 heteroclite mutants.
Protein abundance analysis of the 36 heteroclite genes
Ghaemmaghami et al.  carried out genome-wide analyses of the protein levels in S. cerevisiae. This data were used to compare the abundance of the protein molecules per cell for the 110 selected genes with deletions resulting in a variable morphology. The protein molecules of the 4718 nonessential genes were distributed across a wide range (from 41.1 − 681,000) as were those of the 110 selected genes (from 149 to 378,000; Additional file 11: Figure S8A). To investigate the differences of the abundance of the protein molecules between the 36 heteroclite genes and the others, a Mann–Whitney U-test was performed. We found that no significant differences (Additional file 11: Figure S8B) were detected. This suggested that the morphological abnormalities caused by the deletion of the 36 heteroclite genes were not dependent on protein abundance, and that protein abundance was not a common property in the heteroclite genes.
Assessment of genetic interactions in the 36 heteroclite genes
Gene annotations frequently observed in the 36 heteroclite genes
To identify the gene ontologies (GOs) that were statistically enriched in the 36 heteroclite genes, a multivariate analysis of variance (MANOVA) was applied. We found that of the 13 GOs that were annotated to at least three of the 36 heteroclite genes (Additional file 12: Table S4, see Methods), nine GOs showed a significant correlation (P <0.05 after Bonferroni correction, F-test) with PC1, PC2, or PC4, which differentiated the deletion mutants from the natural strains. This implied that these nine GOs were selected due to the morphological phenotype of the deletion mutants.
The natural yeast strains analyzed were derived from different geographical and ecological origins, and displayed diverse morphological phenotypes . They contained an average of 30,000 single nucleotide polymorphisms (SNPs) and 63 deletion events, with >200 base pairs (bp) per strain . The nonessential gene deletion collection contained a single deletion in every gene, which highlighted the fact that approximately 50% of the deletion mutants had an abnormal morphology compared with their parental strains . Here, we showed a more diverse morphology for the deletion mutants than the natural yeast strains after verifying our procedure with the mosaic progeny and showing the robustness of our analysis for the strain selection. We also identified deletion mutants with morphologies not encountered in the natural yeast population. We propose that these genes are essential gene candidates in nature.
Although natural strains accumulated many nucleotide polymorphisms , their morphology was less diverse than the deletion collection, suggesting a robustness of the cell morphology in the natural yeast strains. Several possible explanations exist for the conserved morphology in the natural strains. First, a cellular mechanism may coordinate an increase in cell size with biosynthetic capacity and nutrient availability . Therefore, natural yeast strains likely have an upper size limit. The deviation of the long axis length in the mother cell was 5.3 − 7.1 and 4.5 − 11.4 μm in natural yeast strains and the deletion collection, respectively. These results showed that natural yeast strains were less diverse in cell size. Second, the morphological parameters that directly affect growth rate may be conserved due to the advantages they confer during the competition to survive. When the ratio of unbudded cells increased, an increase in doubling time was expected . The ratio of unbudded cells was between 13.4% and 71.5% in the natural yeast strains , implying that the unbudded ratio of 71.5% was constitutive. Third, the loss of cell polarity may result in defects in shmoo formation and in the mating process , which could constitute an evolutionary disadvantage. Thus, the evolutionary conservation of morphological traits may link to the survival of the extant budding yeast in nature.
We revealed a higher morphological variation generated by single gene deletions. Yeast deletion mutants with a slow-growth phenotype likely work against natural competition in nature . We found that 22 of the 36 heteroclite deletion mutants had >10% reduction in fitness. In addition to the decreased fitness in the normal medium, the heteroclite deletion mutants were often unable to grow under certain growth conditions, such as high or low temperatures, lack of nutrients, treatment with toxic compounds, and radiation exposure. Survival under severe environment conditions and response to change is essential for preservation of the species . Another possibility is the presence of lesions other than those of vegetative growth. Yeast life cycles are composed of many processes involving different forms and cell shapes . Genes acting on vegetative cellular morphogenesis are sometimes essential for the progression of other yeast life cycles, including germination, mating, meiosis, sporulation, and biofilm formation, which may be confer a competitive advantage in survival. Finally some yeast variants may become disadvantaged during breeding or domestication . Due to their usefulness for alcohol fermentation, Saccharomyces yeast species may acquire their competitive advantage under high ethanol conditions.
Some properties were common in the heteroclite genes. The heteroclite genes frequently displayed a relatively large number of genetic interactions because the deletion mutant genes conferred the most variable morphological phenotypes. We hypothesized that the genes with a high number of genetic interactions acted as a hub in the cellular network , affecting many aspects of cell morphology. Alternatively, the predominant yeast genes involved in essential cellular processes frequently interact with each other and thus influence cell morphology to a major extent. We also ascertained that the genes belonging to the heat-shock response, DNA repair, translation, and protein modification were enriched in the heteroclite mutants. This led to the speculation that these gene functions are essential in nature; e.g., budding requires genes involved in the heat-shock response. The genes involved in DNA repair are important because this function is required for adaptation or survival under exposure to natural radiation. Since exposure to mild stress leads to an increased tolerance for other stresses , these functions may be necessary in preparing for future threats.
High-dimensional morphometric features of Saccharomyces cerevisiae were examined to find that the gene deletion strains had higher morphological variance than the natural strains. Our multivariate analyses revealed gene deletion mutants whose morphologies were not seen in nature. The yeast genes that were practically important in nature were characterized in terms of extent of genetic interaction and specific cellular functions. Although evolution has often been cited in the context of current living species or for those now extinct, this study provided a new insight into the evolution and species conservation of yeast. Further study will be required to fully clarify the evolution of important yeast genes by means of competitive assays with deletion mutants and natural yeast strains.
Strains, culture and morphological analysis
The collection of diploid homozygous deletion strains was purchased from EUROSCARF (the EUROpean Saccharomyces Cerevisiae ARchive for Functional Analysis). The strains were grown in synthetic C medium  at 30°C to logarithmic phase.
To obtain the fluorescent images of the cell wall, actin cytoskeleton, and nuclear DNA, yeast cells were stained triply with fluorescein isothiocyanate-ConA (Sigma Aldrich, St. Louis, MO, USA), rhodamine-phalloidin (Invitrogen, Carlsbad, CA, USA), and 4′,6-diamidino-2-phenylindole (Sigma Aldrich), respectively. At least 200 cells were counted on the fluorescent images from one independent culture, and we visualized sets of images from each homozygous diploid strain. The image sets were processed with the CalMorph software (version 1.3, designed for diploid cells) as described previously . Raw images and datasets are freely available at http://www.yeast.ib.k.u-tokyo.ac.jp/natural_vs_deletion/index.html. All statistical analyses were performed using R software (http://www.r-project.org). To investigate the distribution of the morphological changes, a PCA was performed based on the variance–covariance matrix of the Z-score using the prcomp() function in R.
Statistical models to estimate the effects on cell morphology
To statistically assess the morphological differences of the cells, the GLM, an extension of the normal linear model, was used, which applied not only a Gaussian but also other probability distributions. The models of the probability distributions for the 501 parameters were determined to accommodate the statistical model used in the GLM.
Of the 501 parameters calculated by CalMorph, 220 parameters were coefficients of variation (CV) of their related mean parameters calculated from a single cell trait. The CV parameters depended highly on the mean trait values, and this dependence could be uncoupled by a nonlinear Lowess regression method . To normalize the CV values by uncoupling the dependency, the Lowess regression of the CV values by the mean values was performed using the lowess() function of R with a smooth span of 0.4, as described previously . CVs were assumed to be Gaussian-distributed after the normalization. A further 183 parameters, representing the mean cell morphologies with positive continuous values, were assumed to be gamma-distributed. Another 37 parameters, representing the mean cell morphologies with continuous values ranging from zero to one, were assumed to be beta-distributed. The remaining 61 parameters, representing the ratio of cells in specimen, were assumed to be binomially distributed with overdispersion. The models of the probability distributions and descriptions of the parameters are listed in Additional file 13: Table S5. The single linear model was used for the assessment of the Gaussian, gamma, and beta parameters in the GLM as defined by
where η is link function listed in Additional file 13: Table S5 , y i is the response variable (parameter values), β0 is the intercept, β1 is the slope (fixed effect), x i is the explanatory variable (0 and 1 for the wild-type and the mutants, respectively), and ϵ i is the error. For the BY, RM and their segregants, a single value for each culture was estimated in each of 501 parameters after three replications. For the 36 natural strains and BY4743, a single value for each strain was estimated in each parameter after five and 40 replications, respectively, with three of the 501 parameters being discarded due to a missing value. The Wald test for the maximum likelihood estimation of β1 was used as the Z-score in this study.
Selection of representative diploid gene deletion strains by Mahalanobis distance
To select the homozygous diploid gene deletion strains for the morphological analysis, we calculated the Mahalanobis distance of each of 4718 haploid from a central distribution of wild-type strains. Z-values of 501 parameters were calculated by assuming one of four probability distribution models for each parameter (Additional file 13: Table S5) and were used to calculate the Mahalanobis distance using the mahalanobis() function in R. We selected 100 strains with the highest Mahalanobis distance scores (Additional file 2: Figure S2A) and 20 strains located at the edge of the PC1-PC20 space of 4718 gene deletion strains (Additional file 2: Figure S2B), making a total of 120 strains (Additional file 3: Table S1) covering 7.4% of the total 4718 haploid strains distance. These 120 strains were investigated; the diploid morphology of 110 strains was successfully analyzed, with 10 strains failing to supply any morphological data due to poor growth (Additional file 3: Table S1 and Additional file 2: Figure S2C). In addition, we analyzed the morphology of the next most variable 20 strains with Mahalanobis distance scores (Additional file 3: Table S1) to confirm the robustness of the analyses.
Estimation of phenotypic variance and detection of gene deletion strains beyond the variation of wild yeasts
To estimate the variance between BY and RM as shown in Figure 2B, we used the published data . We calculated the sum of squares of the Z-value after it was centered by the mean values of BY (n =3) and RM (n =3), respectively, and divided by 4 (the degrees of freedom). For the 37 natural strains in Figure 3A, we used the published data . The sum of squares was calculated for each parameter after the Z-values were centered by the mean of the 37 strains, and divided by 36 (the degrees of freedom) not to be affected by the sample size. For the 110 gene deletion strains, the Z-values were centered by the mean of 110 strains, and the sum of the squares was divided by 109. To estimate the equiprobability density ellipses in two-dimensional space as shown in Figures 2D and 3C, the variance–covariance matrices were calculated from the PC scores after centering, as described above, and the probability density of the multivariate normal distribution was estimated for each of 200 × 200 bins on the two-dimensional space of each pair of PCs using the dmvnorm() function in R. The equiprobability density ellipse was drawn using the contour() function in R. To detect the gene deletion strains beyond the variance of the natural strains, the variance of the 37 natural strains was estimated by centering with the mean of BY4743, and a one-sample two-sided test of normal distribution from the mean of BY4743 was applied for each gene deletion strain in each PC. The Mann–Whitney U-test with corrections for both of continuity and ties was performed using wilcox.test() function in R.
Multivariate analysis of variance (MANOVA) to detected gene ontologies (GOs) annotated to genes conferring morphological variation from the natural strains
where the Y phenotypes consist of three PCs (PC1, PC2, and PC4) as a factor of the background strains, and natural and homozygous gene deletion strains with a BY background. χ was a factor of the type of mutation defined by GO, where the wild type and natural strains had no artificial gene deletion (0 as a dummy variable) but where the homozygous gene deletion strains did have a gene deletion (1 as a dummy variable). Then we applied the linear model function (lm() function in R) to the null hypothesis: PC1 + PC2 + PC4 = s (cell morphology was explained only by the background of each strain) and the alternative hypothesis: PC1 + PC2 + PC4 = s + x (cell morphology was explained by background and mutation type). These two hypotheses were assessed by F-test. Finally, each of the 13 GOs (Additional file 12: Table S4) was subjected to a MANOVA with the natural strains and the wild-type BY4743, and the gene deletion strains were annotated to each of the 13 GOs.
We thank Charlie Boone and Michael Costanzo for providing unpublished genetic interaction data, Satoru Nogami for initiating analysis of the morphology data, Joseph Schacherer for comments on the manuscript, and members of the Laboratory of Signal Transduction for helpful discussions. Financial support was provided by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan (24370002 to Y.O.). S.O. was a Research Fellow of the Japan Society for the Promotion of Science.
- Goffeau A, Barrell BG, Bussey H, Davis RW, Dujon B, Feldmann H, Galibert F, Hoheisel JD, Jacq C, Johnston M, Louis EJ, Mewes HW, Murakami Y, Philippsen P, Tettelin H, Oliver SG: Life with 6000 genes. Science. 1996, 274 (5287): 563-567.View ArticleGoogle Scholar
- Schacherer J, Shapiro JA, Ruderfer DM, Kruglyak L: Comprehensive polymorphism survey elucidates population structure of Saccharomyces cerevisiae. Nature. 2009, 458 (7236): 342-345. 10.1038/nature07670.PubMed CentralPubMedView ArticleGoogle Scholar
- Liti G, Carter DM, Moses AM, Warringer J, Parts L, James SA, Davey RP, Roberts IN, Burt A, Koufopanou V, Liti G, Carter DM, Moses AM, Warringer J, Parts L, James SA, Davey RP, Roberts IN, Burt A, Koufopanou V, Tsai IJ, Bergman CM, Bensasson D, O'Kelly MJ, van Oudenaarden A, Barton DB, Bailes E, Nguyen AN, Jones M, Quail MA, Goodhead I, Sims S, Smith F, Blomberg A, Durbin R, Louis EJ: Population genomics of domestic and wild yeasts. Nature. 2009, 458 (7236): 337-341. 10.1038/nature07743.PubMed CentralPubMedView ArticleGoogle Scholar
- Warringer J, Zorgo E, Cubillos FA, Zia A, Gjuvsland A, Simpson JT, Forsmark A, Durbin R, Omholt SW, Louis EJ, Liti G, Moses A, Blomberg A: Trait variation in yeast is defined by population history. PLoS Genet. 2011, 7 (6): e1002111-10.1371/journal.pgen.1002111.PubMed CentralPubMedView ArticleGoogle Scholar
- Ansel J, Bottin H, Rodriguez-Beltran C, Damon C, Nagarajan M, Fehrmann S, Francois J, Yvert G: Cell-to-cell stochastic variation in gene expression is a complex genetic trait. PLoS Genet. 2008, 4 (4): e1000049-10.1371/journal.pgen.1000049.PubMed CentralPubMedView ArticleGoogle Scholar
- Ohya Y, Sese J, Yukawa M, Sano F, Nakatani Y, Saito TL, Saka A, Fukuda T, Ishihara S, Oka S, Suzuki G, Watanabe M, Hirata A, Ohtani M, Sawai H, Fraysse N, Latge JP, Francois JM, Aebi M, Tanaka S, Muramatsu S, Araki H, Sonoike K, Nogami S, Morishita S: High-dimensional and large-scale phenotyping of yeast mutants. Proc Natl Acad Sci U S A. 2005, 102 (52): 19015-19020. 10.1073/pnas.0509436102.PubMed CentralPubMedView ArticleGoogle Scholar
- Levy SF, Siegal ML: Network hubs buffer environmental variation in Saccharomyces cerevisiae. PLoS Biol. 2008, 6 (11): e264-10.1371/journal.pbio.0060264.PubMed CentralPubMedView ArticleGoogle Scholar
- Ho WC, Zhang J: The genotype-phenotype map of yeast complex traits: basic parameters and the role of natural selection. Mol Biol Evol. 2014, 31 (6): 1568-1580. 10.1093/molbev/msu131.PubMed CentralPubMedView ArticleGoogle Scholar
- Yvert G, Ohnuki S, Nogami S, Imanaga Y, Fehrmann S, Schacherer J, Ohya Y: Single-cell phenomics reveals intra-species variation of phenotypic noise in yeast. BMC Syst Biol. 2013, 7: 54-10.1186/1752-0509-7-54.PubMed CentralPubMedView ArticleGoogle Scholar
- Nelder JA, Wedderburn RWM: Generalized linear models. J R Stat Soc A. 1972, 135 (3): 370-384. 10.2307/2344614.View ArticleGoogle Scholar
- Gill DE, Chao L, Perkins SL, Wolf JB: Genetic mosaicism in plants and clonal animals. Annu Rev Ecol Syst. 1995, 26: 423-444. 10.1146/annurev.ecolsys.26.1.423.View ArticleGoogle Scholar
- Nogami S, Ohya Y, Yvert G: Genetic complexity and quantitative trait loci mapping of yeast morphological traits. PLoS Genet. 2007, 3 (2): e31-10.1371/journal.pgen.0030031.PubMed CentralPubMedView ArticleGoogle Scholar
- Ohnuki S, Kobayashi T, Ogawa H, Kozone I, Ueda JY, Takagi M, Shin-Ya K, Hirata D, Nogami S, Ohya Y: Analysis of the biological activity of a novel 24-membered macrolide JBIR-19 in Saccharomyces cerevisiae by the morphological imaging program CalMorph. FEMS Yeast Res. 2012, 12 (3): 293-304. 10.1111/j.1567-1364.2011.00770.x.PubMedView ArticleGoogle Scholar
- Ohnuki S, Enomoto K, Yoshimoto H, Ohya Y: Dynamic changes in brewing yeast cells in culture revealed by statistical analyses of yeast morphological data. J Biosci Bioeng. 2014, 117 (3): 278-284. 10.1016/j.jbiosc.2013.08.005.PubMedView ArticleGoogle Scholar
- Warringer J, Ericson E, Fernandez L, Nerman O, Blomberg A: High-resolution yeast phenomics resolves different physiological features in the saline response. Proc Natl Acad Sci U S A. 2003, 100 (26): 15724-15729. 10.1073/pnas.2435976100.PubMed CentralPubMedView ArticleGoogle Scholar
- Ghaemmaghami S, Huh WK, Bower K, Howson RW, Belle A, Dephoure N, O’Shea EK, Weissman JS: Global analysis of protein expression in yeast. Nature. 2003, 425 (6959): 737-741. 10.1038/nature02046.PubMedView ArticleGoogle Scholar
- Costanzo M, Baryshnikova A, Bellay J, Kim Y, Spear ED, Sevier CS, Ding H, Koh JL, Toufighi K, Mostafavi S, Prinz J, St Onge RP, VanderSluis B, Makhnevych T, Vizeacoumar FJ, Alizadeh S, Bahr S, Brost RL, Chen Y, Cokol M, Deshpande R, Li Z, Lin ZY, Liang W, Marback M, Paw J, San Luis BJ, Shuteriqi E, Tong AH, van Dyk N, et al: The genetic landscape of a cell. Science. 2010, 327 (5964): 425-431. 10.1126/science.1180823.PubMedView ArticleGoogle Scholar
- Goranov AI, Gulati A, Dephoure N, Takahara T, Maeda T, Gygi SP, Manalis S, Amon A: Changes in cell morphology are coordinated with cell growth through the TORC1 pathway. Curr Biol. 2013, 23 (14): 1269-1279. 10.1016/j.cub.2013.05.035.PubMedView ArticleGoogle Scholar
- Baroni MD, Martegani E, Monti P, Alberghina L: Cell size modulation by CDC25 and RAS2 genes in Saccharomyces cerevisiae. Mol Cell Biol. 1989, 9 (6): 2715-2723.PubMed CentralPubMedView ArticleGoogle Scholar
- Evangelista M, Blundell K, Longtine MS, Chow CJ, Adames N, Pringle JR, Peter M, Boone C: Bni1p, a yeast formin linking cdc42p and the actin cytoskeleton during polarized morphogenesis. Science. 1997, 276 (5309): 118-122. 10.1126/science.276.5309.118.PubMedView ArticleGoogle Scholar
- Thatcher JW, Shaw JM, Dickinson WJ: Marginal fitness contributions of nonessential genes in yeast. Proc Natl Acad Sci U S A. 1998, 95 (1): 253-257. 10.1073/pnas.95.1.253.PubMed CentralPubMedView ArticleGoogle Scholar
- Causton HC, Ren B, Koh SS, Harbison CT, Kanin E, Jennings EG, Lee TI, True HL, Lander ES, Young RA: Remodeling of yeast genome expression in response to environmental changes. Mol Biol Cell. 2001, 12 (2): 323-337. 10.1091/mbc.12.2.323.PubMed CentralPubMedView ArticleGoogle Scholar
- Knop M: Yeast cell morphology and sexual reproduction - a short overview and some considerations. Cr Biol. 2011, 334 (8–9): 599-606.View ArticleGoogle Scholar
- Sicard D, Legras JL: Bread, beer and wine: yeast domestication in the Saccharomyces sensu stricto complex. C R Biol. 2011, 334 (3): 229-236. 10.1016/j.crvi.2010.12.016.PubMedView ArticleGoogle Scholar
- Zakrzewska A, van Eikenhorst G, Burggraaff JEC, Vis DJ, Hoefsloot H, Delneri D, Oliver SG, Brul S, Smits GJ: Genome-wide analysis of yeast stress survival and tolerance acquisition to analyze the central trade-off between growth rate and cellular robustness. Mol Biol Cell. 2011, 22 (22): 4435-4446. 10.1091/mbc.E10-08-0721.PubMed CentralPubMedView ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.