Strength of selection potentiates distinct adaptive responses in an evolution experiment with outcrossing yeast

Selection intensity is expected to influence the magnitude and genetic architecture of adaptive responses, yet it is rarely evaluated as a standalone variable in experimental evolution studies. Here, we evolved outcrossing populations of Saccharomyces cerevisiae for ~200 generations across a spectrum of environmental stress from zero to moderate to high ethanol exposure, to examine how genomic responses vary with stress intensity. Across treatments, adaptation proceeded through many subtle allele and haplotype frequency shifts rather than large changes at single loci, consistent with a highly polygenic response. At loci associated with ethanol adaptation, the high stress treatment led to larger allele frequency changes compared to the moderate or no ethanol stress treatments, with the genomic architecture of adaptation becoming increasingly polygenic as selection intensity decreased. Moderate and high stress conditions engaged partially distinct biological pathways, indicating that selection intensity shapes both the magnitude and targets of adaptive change. Within this stress continuum, we also observed substantial, ongoing adaptation in control populations despite extensive prior domestication. Many alleles associated with this adaptation showed reduced or absent responses under ethanol stress, consistent with antagonistic pleiotropy. Consequently, laboratory adaptation can represent a major component of evolutionary change and may confound treatment-specific inferences when not explicitly accounted for. Broadly, our results demonstrate that selection intensity structures adaptive responses in experimental evolution and that continued laboratory adaptation remains an important force in these studies. Our findings underscore the importance of clearly defined controls and careful consideration of selection intensity when interpreting or comparing across experimental evolution studies.

Methods
Data is from an evolution experiment with outcrossing yeast populations derived from the "S12" base population, which originated from 12 haploid founder strains. Sixty replicate populations were established: 20 control (C), 20 moderate ethanol stress (M, 6% ethanol), and 20 high ethanol stress (H, 10% ethanol). Populations underwent weekly outcrossing and batch culture for 15 cycles, corresponding to approximately 200 generations. Growth rate assays were conducted at the end of the experiment to quantify adaptation by measuring optical density over 48 hours in control and treatment-specific media. DNA was extracted from pooled populations at three timepoints (cycles 1, 7, and 15) and sequenced using Illumina PE150. Raw reads were aligned to the S. cerevisiae S288C reference genome, and SNPs were called using GATK. Variant tables were filtered to retain high-confidence polymorphic sites. Candidate SNPs responding to selection were identified using a modified Cochran-Mantel-Haenszel (CMH) test, and selection coefficients were estimated genome-wide using Bait-ER. Haplotype frequencies were estimated with a sliding-window haplotype caller to assess the contributions of rare and founder-specific variants. More specific details on population maintenance, phenotyping, and sequencing can be found in the manuscript entitled “Increased time sampling in an evolve-and-resequence experiment with outcrossing Saccharomyces cerevisiae reveals multiple paths of adaptive change”. These datasets include growth measurements, SNP tables, CMH results, selection coefficient estimates, and haplotype frequency data for ancestral and evolved populations across treatments and timepoints (see ReadMe file for details).