Back

Domestication and deep lineage divergence define two discrete trajectories in Kluyveromyces marxianus

Moreno-Hernandez, J. M.; Smets, J.; Torres-Machorro, A. L.; Garcia-Ortega, L. F.; Avelar-Rivas, J. A.; Ortiz- Merino, R. A.; Kirchmayr, M. R.; Muszewska, A.; Morales, L.; Verstrepen, K.; DeLuna, A.; Mancera, E.

2026-04-29 evolutionary biology
10.64898/2026.04.28.721375 bioRxiv
Show abstract

Although genomics has greatly improved our ability to detect population structure and genetic divergence, it remains difficult to determine when diversity within a recognized species reflects continuous variation and when it marks the emergence of discrete evolutionary lineages. In yeasts, population-genomic studies often reveal overlapping signatures of environmental association, admixture, and domestication, making this distinction especially challenging. By analyzing 178 globally sampled genomes from Kluyveromyces marxianus, an emerging yeast of biotechnological interest, we show that diversity within this species resolves into two contrasting evolutionary trajectories that do not map onto broad geographic origin. Most industrial and clinical isolates belong to a lineage marked by gene loss, aneuploidy, and reduced fertility, consistent with genomic hallmarks of domestication. In contrast, isolates confined to traditional agave fermentations in Mexico and South Africa form a highly distinct lineage, separated from the other lineages by up to 4.8% genome-wide nucleotide divergence, widespread reciprocal monophyly, no detectable recent gene flow, and reduced mating compatibility despite co-occurrence with other lineages. Together, these results show that within a single yeast species, domestication-associated genome remodeling and deep lineage divergence can define distinct trajectories, with implications for strain selection in biotechnology.

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.