כתבה
arXiv cs.LG ·
PIT-GCL: חידוש באיתור התגובה של חלבונים
PIT-GCL: Protein Interaction using Topological Graph Contrastive Learning
איתור התגובה של חלבונים: חידוש בשיטה לאיתור התגובה של חלבונים באמצעות למידת מודלים עם תכונות טופולוגיות.
תקציר מקורי באנגליתarXiv:2610.04850v2 Announce Type: replace Abstract: Protein binding prediction is central to target identification, therapeutic binder design, and large scale screening, yet remains challenging because binding depends on sequence, three dimensional geometry, and global structural organization. Recent folding models such as AlphaFold3 and Boltz-2 have substantially improved structure prediction, but their confidence outputs (pLDDT, pTM, ipTM) are not specifically designed for binary binding prediction, and dedicated structure aware predictors often require bound complex structures that are unavailable at screening scale. We introduce PIT-GCL, a dual tower structure aware framework that encodes each protein independently from its amino acid sequence, C{\alpha} point cloud, and a global persi
קרא במקור המקורי
arxiv.org
פתח כתבה מקורית