Evolutionary genomics of transposable elements in Saccharomyces cerevisiae

Saccharomyces cerevisiae is one of the premier model systems for studying the genomics and evolution of transposable elements. The availability of the S. cerevisiae genome led to many insights into its five known transposable element families (Ty1-Ty5) in the years shortly after its completion. However, subsequent advances in bioinformatics tools for analysing transposable elements and the recent availability of genome sequences for multiple strains and species of yeast motivates new investigations into Ty evolution in S. cerevisiae. Here we provide a comprehensive phylogenetic and population genetic analysis of Ty families in S. cerevisiae based on a reannotation of Ty elements in the S288c reference genome. We show that previous annotation efforts have underestimated the total copy number of Ty elements for all known families. In addition, we identify a new family of Ty3-like elements related to the S. paradoxus Ty3p which is composed entirely of degenerate solo LTRs. Phylogenetic analyses of LTR sequences identified three families with short-branch, recently active clades nested among long branch, inactive insertions (Ty1, Ty3, Ty4), one family with essentially all recently active elements (Ty2) and two families with only inactive elements (Ty3p and Ty5). Population genomic data from 38 additional strains of S. cerevisiae show that elements present in active clades are predominantly polymorphic, whereas most of the inactive elements are fixed. Finally, we use comparative genomic data to provide evidence that the Ty2 and Ty3p families have arisen in the S. cerevisiae genome by horizontal transfer. Our results demonstrate that the genome of a single individual contains important information about the state of TE population dynamics within a species and suggest that horizontal transfer may play an important role in shaping the diversity of transposable elements in unicellular eukaryotes.

💡 Research Summary

The authors revisit the transposable element (TE) landscape of the model yeast Saccharomyces cerevisiae by applying modern bioinformatic pipelines to the S288c reference genome and by integrating population‑genomic data from 38 additional strains. Using a custom RepeatMasker library that includes the five canonical Ty families (Ty1‑Ty5) plus the S. paradoxus Ty3p element, they re‑annotated the genome with the REANNOTATE tool, manually curating fragmented and nested insertions. This effort uncovered 483 Ty insertions—a 46 % increase over the 331 copies reported by Kim et al. (2008). The extra copies are largely degenerate solo LTRs; only a modest amount of additional DNA (~30 kb) is contributed by full‑length elements, confirming that most Ty copies in yeast are fragmented.

A notable discovery is a previously unreported Ty3‑like family, termed Ty3p, composed entirely of solo LTRs in S. cerevisiae. Phylogenetic analyses of LTR sequences were performed separately for each family using maximum‑likelihood (RAxML) and Bayesian (MrBayes) methods. The resulting trees reveal distinct evolutionary patterns: Ty1, Ty3, and Ty4 contain both short‑branch (recently active) and long‑branch (ancient, inactive) clades, indicating ongoing transposition alongside a legacy of older insertions. Ty2 is dominated by short‑branch clades, suggesting that most Ty2 copies are still part of an active transposition wave. Ty3p and Ty5 consist exclusively of long‑branch, inactive elements.

To connect phylogenetic inference with actual population dynamics, the authors examined resequencing data from the Saccharomyces Genome Resequencing Project (SGRP). For each of the 483 annotated insertions they extracted the corresponding region in the 38 strains, manually scoring presence, absence, or missing data. Insertions belonging to recently active clades are predominantly polymorphic (present in some strains, absent in others), whereas the majority of inactive insertions are fixed across the panel. This concordance validates the use of branch length as a proxy for recent activity. A small subset of young insertions occurs at high frequency, hinting at possible positive selection, perhaps due to adaptive effects on gene regulation.

The authors also investigated the origins of Ty2 and Ty3p. By constructing multi‑species alignments of LTRs from the entire Saccharomyces sensu stricto clade and generating super‑family phylogenies (Ty1/Ty2 and Ty3/Ty3p), they observed that S. cerevisiae Ty2 clusters with sequences from other species rather than with its own older Ty1 copies, and that Ty3p groups with S. paradoxus Ty3p rather than with native Ty3. These patterns, together with the lack of intermediate forms, support horizontal transfer events: Ty2 appears to have entered the S. cerevisiae lineage from another Saccharomyces species, and Ty3p likely arrived from S. paradoxus. The timing of these transfers differs, suggesting multiple, independent acquisition events.

Overall, the study demonstrates that a single, well‑annotated reference genome can reveal much about TE population dynamics, but full insight requires comparative data from many strains and related species. The increased copy number, identification of a new Ty family, the clear separation of active versus inactive clades, and the evidence for horizontal transfer collectively reshape our understanding of TE evolution in yeast. The work underscores the importance of integrating high‑throughput sequencing, robust annotation pipelines, and phylogenetic frameworks to dissect the complex evolutionary forces—selection, drift, recombination, and horizontal acquisition—that shape transposable element landscapes in eukaryotes.