The abstract is read at the publisher; the summary is JClub's.
PTM-BDL predicts log-transformed half-maximal inhibitory concentration and resistance with internal AUROCs of 0.785–0.946 across BCR-ABL, HDAC/EP300, and EGFR/ERBB2 settings while preserving site-level attributions.
- Why it matters: Fixed molecular features miss how treatment-associated signalling depends on which residues are modified, limiting how well drug-response models can be linked to pharmacodynamic measurements.
- What they did: The framework combines typed PTM states with protein sequence, structure-derived, and drug chemical features, using integrated gradients for site interpretation and evaluating AUROC on retrospective PTM profiles.
- The result: In K562/CML, adding PTM inputs increased AUROC by 0.0268 on average with gains in all five folds; CRKL Y207 and STAT5A Y694 ranked highest, and PTM attribution was stronger for dasatinib than cytotoxic contexts.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.