The abstract is read at the publisher; the summary is JClub's.
HieraPTM-HieraFuncPhos predicts functional phosphosites with biophysically grounded mechanical patterns at HSP90AA1 T36 and T115, matching published NMR observations.
- Why it matters: PTM site recognition and functional phosphosite prediction are often modeled separately, limiting how well recognition knowledge transfers into functional prioritization.
- What they did: The framework uses a hierarchical, cross-task design: HieraPTM combines local sequence, protein language model features, and 3D structure for multi-type PTM recognition, while HieraFuncPhos adapts representations and retrieves local and distal mechanical evidence conditioned on functional context.
- The result: Across eight PTM tasks, HieraPTM reaches macro-averaged AUROC 0.9237 and AUPRC 0.8425, and HieraFuncPhos attains AUROC 0.8274 and AUPRC 0.8188, outperforming existing comparison methods.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.