AI Tool auto-optimize Speeds Up Picard MarkDuplicates 2-9x with Byte-Identical Output
jmschreiber91 · x · 2026-10-30
auto-optimize, developed by the Programmable Genomics Lab at UMass Chan Medical School, submitted a PR to the bioinformatics tool Picard: MarkDuplicates' median wall time on a 22M-pair DNase-seq BAM at 8 cores drops from 234.2s to 25.05s — a 2.0x to 9.4x speedup across 56 public BAM files.
Correctness was verified at scale: 28,294 comparisons against master across 2,253 inputs from 50 assays (BAM/SAM/CRAM), including deliberately corrupted BGZF blocks, truncated files, and other edge cases, with byte-identical output. The team is also optimizing minimap2 and other genomics tools.
Related event: AI auto-optimize tool speeds up classic bioinformatics tools up to 10x(3 posts)→
More from Research
- After formalizing Fermat's Last Theorem, community scales the approach to 2,012 operations research papers in Lean 4 — ChrSzegedy · 2026-10-30
- Datology AI builds synthetic pretraining data with measurable knowledge cutoff guarantees — josh_wills · 2026-10-30
- PlasmaAI Announces Proof of Erdős #173, a Geometry Conjecture From 1973 — kevrussell · 2026-10-30
- Study of 26,811 Chinese students: AI lifted homework scores 18% but cut exam scores 20% — ValerioCapraro · 2026-10-30
- NTv3 genome foundation model released with 15-species function prediction and regulatory sequence design — anshulkundaje · 2026-10-30
- Persona neurons fire hardest on chat-template tokens, carrying 25–95% of total activation — maxsloef · 2026-10-30