Arc Institute uses genomic language model to map 376M prokaryotic noncoding sequences into 5.14M clusters
tedgreenwald · x · 2026-10-23
Arc Institute's Patrick Hsu team released a preprint introducing COMPASS, a genomic language model approach to systematically map the noncoding 'dark matter' of prokaryotic genomes.
- Embeddings from 1.3 terabases of prokaryotic genomes/metagenomes organized 376 million intergenic regions into 5.14 million clusters
- Nominated 20,149 clusters of putative structured RNAs with candidate protein partners; recovered CRISPR, TIGR and VIPR arrays plus cognate effector proteins from sequence alone
- Identified bridge RNA variants with longer guide recognition sequences than any characterized ortholog, and deployed teams of AI agents to mine the database for divergent riboswitches and ribozymes with functional hypotheses
- A public database is released to enable human-agent molecular discovery
More from Research
- 1-gram wireless mesoscope images cortex-wide activity in freely moving mice — KordingLab · 2026-10-23
- Reversible computing veteran poses open problem: universal BARCS primitives from unbiased Josephson junctions — MikePFrank · 2026-10-23
- The rise of AI-assisted peer review: reviewers are making it into the appendix — abursuc · 2026-10-23
- Anthropic uses Claude to formally verify security properties in 100+ open-source libraries at RISC-V assembly level — AnthropicAI · 2026-10-23
- Researchers test 9 strategies to pick LLM teams, beating random selection by up to 36.1% — shangbinfeng · 2026-10-23
- Transformer pretraining without backprop drops from 1.76 hours to 1.6 seconds in 10 days — mayfer · 2026-10-23