Supplementary Fig. 10: At RH-seq hit loci, the effect of allelic variation is conserved across a given species and sequence divergence from S. paradoxus is a common feature of S. cerevisiae strains.
From: Genetic dissection of interspecific differences in yeast thermotolerance

a, Each pair of bars reports the growth efficiency of an S. cerevisiae DBVPG1373 strain harboring the allele of ESP1 from the indicated strain of S. paradoxus, relative to the analogous quantity for wild-type S. cerevisiae DBVPG1373; the heights of the dark and light bars report the mean relative efficiency at 39 °C (n = 4–18 cultures) and 28 °C (n = 22–33 cultures), respectively. Individual measurements are reported as circles. b, Data and symbols are as in a, except that each bar reports results from an S. paradoxus Z1 strain harboring the allele of APC1 from the indicated strain of S. cerevisiae, relative to wild-type S. paradoxus Z1. *P ≤ 0.034 in a one-sample, one-tailed t test (39 °C) or a one-sample, two-tailed t test (28 °C) for a difference in efficiency between the indicated allele replacement strain and 1. The provenance of each strain is as follows: Z1, oak bark, UK; N17, oak exudate, Russia; IFO1804, oak bark, Japan; DBVPG1373, soil, Netherlands; DBVPG1788, soil, Finland; YPS128, soil, USA; DBVPG6044, bili wine, West Africa. c, Each row reports a comparison of the sequences of RH-seq hit loci against a genomic null. S. cer branch length is the number of sequence substitutions along the lineage leading to S. cerevisiae, in a phylogenetic tree inferred from Saccharomyces species type strains. Dxy/length is the average number of differences between the S. paradoxus type strain and a strain randomly chosen from the S. cerevisiae wine/European population, normalized by gene length. The first and second columns report the average of the indicated statistic across the eight RH-seq hit loci and across sets of eight loci randomly resampled from the genome, respectively. The third column reports the empirical P value from a test for an elevated value of the indicated statistic relative to the resampling null. See Supplementary Table 1 for exact P values and sample numbers.