Popular repositories Loading

1. genome-annotation-db genome-annotation-db Public

   Normalised relational database over Ensembl gene annotation, with a validating loader and a benchmark measuring what indexes are actually worth

   Python

2. chem-benchmark-audit chem-benchmark-audit Public

   How much of molecular ML survives an honest split? Random splits leak 67% of scaffolds. Closing the leak costs every model about 0.13 RMSE, and no deep-model advantage was established. 45 cells ove…

   Python

3. dhfr-campaign dhfr-campaign Public

   Property-matched decoys cut a DHFR screen's property-only AUC from 0.913 to 0.841 - not to 0.5. Matching reduces decoy bias without removing it.

   Python

4. fpsearch-rs fpsearch-rs Public

   Thresholded top-k Tanimoto search over binary fingerprints in Rust. A popcount-sorted, memory-mapped index with the BitBound bound gives a top-10 search over 2.85M ChEMBL molecules at Tanimoto 0.95…

   Rust

5. peptide-binder-design peptide-binder-design Public

   The same sequence filter scores AUC 0.684 against naive controls but only 0.569 against composition-preserving scrambles, and recall at a matched control rate falls from 27.4% to 6.3%. The choice o…

   Python

6. sc-spatial-pipeline sc-spatial-pipeline Public

   How much of a spatial transcriptomics result is the pipeline? One preprocessing choice at a time on Visium mouse brain gives median ARI 0.83, and with clusters matched between runs the neighbourhoo…

   Python