Repository navigation
Tags
Tags
chore(main): release aberrant-expression 0.1.1 (#38) 🤖 I have created a release *beep* *boop* --- ## [0.1.1](aberrant-expression-v0.1.0...aberrant-expression-v0.1.1) (2026-10-11) ### Bug Fixes * **deps:** require polars-bio 0.36.1 ([#36](#36)) ([8fe3fba](8fe3fba)) * **nmd_scanner:** drop the unused yaml import of the annotation script ([84a7daf](84a7daf)) * **nmd_scanner:** take the NMD feature flags from all transcripts ([#39](#39)) ([84a7daf](84a7daf)) * **utils:** let check_is_fitted pass on a NullRegressor ([84a7daf](84a7daf)) * **utils:** let scikit-learn see fitted AbExp model wrappers ([84a7daf](84a7daf)) * **utils:** let sklearn.clone copy AbExp model wrappers around a ([84a7daf](84a7daf)) * **workflow:** tell polars the memory limit of the job ([84a7daf](84a7daf)) --- This PR was generated with [Release Please](https://github.com/googleapis/release-please). See [documentation](https://github.com/googleapis/release-please#release-please). Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
chore(main): release aberrant-expression 0.1.0 (#33) 🤖 I have created a release *beep* *boop* --- ## [0.1.0](aberrant-expression-v0.0.1...aberrant-expression-v0.1.0) (2026-10-08) ### ⚠ BREAKING CHANGES * merge the AbExp packages into aberrant-expression ([#18](#18)) ### Features * **absplice2:** score Pangolin with abexp.pangolin ([#19](#19)) ([31313e6](31313e6)) ### Code Refactoring * merge the AbExp packages into aberrant-expression ([#18](#18)) ([0235b79](0235b79)) --- This PR was generated with [Release Please](https://github.com/googleapis/release-please). See [documentation](https://github.com/googleapis/release-please#release-please). Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
test(absplice2): pin absplice2_dna to the upstream AbSplice2 scripts tl;dr: 12 tests run absplice2_dna on made-up inputs, one constellation each, with a stand-in for the AbSplice2 model. make_expected.py computes their expected rows with the AbSplice2 scripts at a30120f. - Constellations: a gain 2 bp from a minus-strand site (3 bp does not match), gain and loss both matched, gain and loss of equal size, a gain that rounds to 0, rounding half to even, a gain of 0.9 above the cap, a deletion, tied rows in both orders, a PAR gene on chrX and chrY, two tissues, and a gene that only MMSplice scored. - The stand-in model is a logistic function of the features, read by position, so the tests also pin the feature order. - Of tied rows, the upstream scripts keep the first in input order. make_expected.py prints both orders; absplice2_dna keeps the same row in both. - packages.yml runs the new area tests/absplice2 with the extra absplice2. Belongs-to: #19