Population genomics of domestic and wild yeasts

Abstract

Since the completion of the genome sequence of Saccharomyces cerevisiae in 1996 (refs 1, 2), there has been a large increase in complete genome sequences, accompanied by great advances in our understanding of genome evolution. Although little is known about the natural and life histories of yeasts in the wild, there are an increasing number of studies looking at ecological and geographic distributions, population structure and sexual versus asexual reproduction. Less well understood at the whole genome level are the evolutionary processes acting within populations and species that lead to adaptation to different environments, phenotypic differences and reproductive isolation. Here we present one- to fourfold or more coverage of the genome sequences of over seventy isolates of the baker's yeast S. cerevisiae and its closest relative, Saccharomyces paradoxus. We examine variation in gene content, single nucleotide polymorphisms, nucleotide insertions and deletions, copy numbers and transposable elements. We find that phenotypic variation broadly correlates with global genome-wide phylogenetic relationships. S. paradoxus populations are well delineated along geographic boundaries, whereas the variation among worldwide S. cerevisiae isolates shows less differentiation and is comparable to a single S. paradoxus population. Rather than one or two domestication events leading to the extant baker's yeasts, the population structure of S. cerevisiae consists of a few well-defined, geographically isolated lineages and many different mosaics of these lineages, supporting the idea that human influence provided the opportunity for cross-breeding and production of new combinations of pre-existing variations.

Fig.1. Saccharomyces phylogenomics

NJ trees based on SNP differences of a, S. cerevisiae and S. paradoxus strains sequenced in this project, using S. mikatae, S. kudriavzevii and S. bayanus as outgroups; b, Close-up of the European S. paradoxus, with UK isolates highlighted in violet; c, S. cerevisiae strains with clean lineages highlighted in grey, with colour indicating source (name) and geographic origin (dots).

Fig. 2. Saccharomyces population structure

a, Inference of population structure using STRUCTURE on S. paradoxus (markers: 7544 SNPs with >30 strains passing neighbourhood quality standard, NQS), assuming K=6 subpopulations and correlated allele frequencies, linkage model based on marker distances in basepairs, 15000 iteration burn in, and 5000 iterations of sampling. Each mark on the x axis represents one strain, and the blocks of colour represent the fraction of the genetic material in each strain assigned to each cluster. b, As a, but for S. cerevisiae (markers: 3413 SNPs with >30 strains passing NQS)

Fig. 3. Population genomics: variation and selection

a, Linkage disequilibrium as a function of distance averaged over 1kb. Insets show the decline in linkage disequilibrium over the first 10kb. Details shown in Table S7. b, Derived allele frequencies of SNPs in coding regions. Amino acid changing SNPs (‘a’) show an excess of low frequencies compared to synonymous SNPs (‘s’). Synonymous SNPs in genes with strong codon bias (‘s*’) are in excess at low and high frequencies. SNPs that create stop codons (‘create stop’) show skew to low frequencies. Inset is the number of mutations occurring over the length of the protein, exceeding three standard deviations from the mean in the C-terminus. c, Distribution of sizes of indel polymorphisms in coding regions. High frequency indels (>10%, red) more often occur in multiples of 3 than low frequency indels (grey). Inset is as for b. d, Frequency distribution of indels in coding regions. Out of frame indels (grey) show excess at low frequencies relative to in frame indels (unfilled). The proportion of out of frame indels decreases as frequency increases. Error bars represent the standard error of the proportion.

Fig. 4. Saccharomyces phenotype variation

A selection of growth phenotypes for S. cerevisiae and S. paradoxus strains in different environments and drugs. The complete set of lag, rate and density phenotypes in 67 environments is displayed in Fig. S9. Phenotypes were quantified using high-resolution micro-cultivation measurements of population density. Strain (n=2) doubling time (rate) phenotypes in relation to the S288c derivative BY4741 (n=20) are displayed. Green = poor growth, red = good growth. Hierarchical clustering of phenotypes was performed using a centered Pearson correlation metric and average linkage mapping. Blue = S. paradoxus, pink = S. cerevisiae, grey = S. bayanus isolate CBS7001.