AI-rewritten MACS3 peak caller hits 40-60x speedup, validated on 2TB genomics test suite
jmschreiber91 · x · 2026-10-30
The author describes his team's AI-driven rewrite of bioinformatics tooling: the biggest improvement so far is the MACS3 peak caller, which saw a roughly 40-60x speedup on common tasks and has already been merged and released by @fooliu.
He stresses validation: unit tests are limited, so the final step is always the Genomics Test Suite — about 2TB of genomics experiments spanning quality, assay, endedness, fragment length and size, plus intentionally malformed versions of every dataset to check the tool also fails in the same way.
More from coding & agent
- Veteran dev pitches free Austin LangChain AI agents meetup on Nov 2 — colinmcnamara · 2026-10-30
- Skyworks cuts root cause analysis from 20 days to 10 with LangGraph agents — LangChain · 2026-10-30
- 73-entry list sorts agent infra: which MCP tools an agent can provision itself vs needs a human — Total-Reasonable · 2026-10-30
- YC-backed Turnstone turns your existing AI subscriptions into an agent team that auto-switches when limits run out — ycombinator · 2026-10-30
- The simple self-testing loop behind every RL data company, per practitioners — teodorio · 2026-10-30
- 7 AI models all faked completion reviewing the same 111k-line codebase — a postmortem — rseroter · 2026-10-30