LLM Agents Reconcile Five Databases Into Open Human Surfaceome Atlas
anshulkundaje · x · 2026-10-21
Surfaceome v1.3.0: An Atlas Built by LLM Agents
Roughly 10–20% of human proteins reach the cell surface, yet over 65% of approved drugs target them — accessibility lets drugs act without entering the cell. Five public databases (UniProt, GO Cellular Component, Human Protein Atlas, Cell Surface Protein Atlas, SURFY) each catalog the surfaceome with different definitions, disagreeing more than they agree.
The Deliverome Project's Surfaceome atlas uses an LLM triage agent to score every protein-coding human gene against the union of the five sources, then a deep-dive agent assembles per-gene evidence records: surface-localization methods, antibody validation, isoform topology, accessibility caveats.
Results are tiered: Canonical (1,783 high-confidence proteins), Likely (1,252 broader set), plus cell-state and cell-type sub-buckets. Fully open access, evidence-cited, agent-readable, with filter definitions exposed via API.
More from Research
- Seqlet caller follow-up: credits for DeepLIFT contributors — jmschreiber91 · 2026-10-21
- Center for AI Safety Publishes Full Syllabus for Intro to ML Safety Course — ArtificialOther · 2026-10-21
- Seqlet caller gets 22x speedup: 167,750 loci drop from 7.8s to 0.35s — jmschreiber91 · 2026-10-21
- X trends: SophontAI's Holon-S1 predicts gene activity from cancer slides; decision models rise — iScienceLuvr · 2026-10-21
- AI auto-optimization speeds genomics code 22x: extract_loci from 7.8s to 0.35s — jmschreiber91 · 2026-10-21
- NEJM AI study: LLM-recommended care costs up to 10x more than physician orders — dvir_a · 2026-10-21