Cross-architecture ensembling of DNA foundation models improves the precision and stability of chimera detection in long-read metagenomic bins
Abstract
Motivation: Chimeric metagenome-assembled genomes (MAGs) that pool DNA from multiple organisms contaminate downstream analyses. Marker-gene tools such as CheckM2 miss low-level chimerism, and DNA foundation models have been proposed as a sequence-composition alternative, but whether large autoregressive models (Evo2, 7B parameters) outperform smaller contrastive models (DNABERT-S, 117M) has not been rigorously tested.
Lifecycle
- biorxiv v1 2026-07-07 source ↗
Discussion
Moderated digest of third-party discussion on Bluesky — substantive endorsement and critique.
-
bioRxiv Bioinfo @biorxiv-bioinfo.bsky.social · 4920 followers neutral
Cross-architecture ensembling of DNA foundation models improves the precision and stability of chimera detection in long-read metagenomic bins https://www.biorxiv.org/content/10.64898/2026.07.02.735979v1
-
AI x Bio Discovery @aixbiobot.bsky.social · 787 followers neutral
Cross-architecture ensembling of DNA foundation models improves the precision and stability of chimera detection in long-read metagenomic bins [new]
-
bioRxivpreprint @biorxivpreprint.bsky.social · 8895 followers neutral
Cross-architecture ensembling of DNA foundation models improves the precision and stability of chimera detection in long-read metagenomic bins https://www.biorxiv.org/content/10.64898/2026.07.02.735979v1