Back

A haploid wild yeast resource for exploring the natural ecology of Saccharomyces cerevisiae

Yang, C.-J.; Yeh, Y.-C.; Hsiao, C.; Liu, Y.-H.; Lu, M. R.; Liu, Y.-C.; Liti, G.; Tsai, I. J.; Ke, H.-M.

2026-02-16 genetics
10.64898/2026.02.13.705822 bioRxiv
Show abstract

Saccharomyces cerevisiae predominantly exists as diploid cells in nature, a life-cycle feature that limits classical genetic analyses of wild populations. Here, we establish a stable haploid strain collection derived from 33 diverse Taiwanese S. cerevisiae isolates through targeted disruption of the HO endonuclease gene. This resource spans predomesticated Asian wild lineages and enables systematic analyses of mating compatibility, reproductive isolation, and ecological trait variation. Although all pairwise hybridizations formed zygotes, many produced low spore viabilities, revealing strong postzygotic barriers. Genome analyses show that reduced hybrid fertility is primarily associated with chromosomal inversions and inter-chromosomal rearrangements rather than sequence divergence, indicating that structural variation maintains lineage separation despite geographic coexistence. Phenotypic profiling uncovered marked ecological differentiation, with the most diverged TW1 lineage favoring cooler growth conditions and a naturally occurring hybrid exhibiting heterosis with expanded thermal tolerance. While most wild strains grew poorly on maltose, two anthropogenically associated strains displayed enhanced maltose utilization linked to functional MAL regulatory alleles and maltose-specific transporters. Together, these findings demonstrate how structural genomic variation and metabolic gene divergence drive ecological and reproductive divergence in wild S. cerevisiae and establish this haploid collection as a platform for studying yeast evolution in nature.

Published in iScience · training set

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.