{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T11:25:34Z","timestamp":1787570734935,"version":"build-2736575974"},"reference-count":29,"publisher":"Oxford University Press (OUP)","issue":"Supplement_2","license":[{"start":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T00:00:00Z","timestamp":1787529600000},"content-version":"vor","delay-in-days":23,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100004761","name":"Natural Science Foundation of Hunan Province of China","doi-asserted-by":"crossref","award":["2025JJ20062"],"award-info":[{"award-number":["2025JJ20062"]}],"id":[{"id":"10.13039\/501100004761","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["U23A20321"],"award-info":[{"award-number":["U23A20321"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62272490"],"award-info":[{"award-number":["62272490"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,8,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Motivation<\/jats:title>\n                    <jats:p>Proteolysis-targeting chimeras (PROTACs) enable targeted protein degradation by recruiting an E3 ubiquitin ligase to a protein of interest (POI) and forming a ternary complex. Despite their therapeutic promise, rational PROTAC design remains challenging, as degradation efficacy depends on subtle and highly structure-dependent interactions among the POI, the E3 ligase, and the bifunctional molecule.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>We propose Pocket-PROTACs, a pocket-aware attention-based framework for predicting PROTAC-induced protein degradation from a triplet of POI, E3 ligase, and PROTAC. Pocket-PROTACs encodes protein sequences using a pre-trained protein language model and represents PROTACs with a geometry-aware graph neural network over an ensemble of three-dimensional conformers. Both POI\u2013PROTAC and E3 ligase\u2013PROTAC interactions are explicitly modeled through a residue\u2013atom cross-attention mechanism that captures fine-grained interaction patterns. To improve model interpretability, we introduce a pocket-aware module that incorporates structural context to guide residue-level relevance estimation, enabling multi-level attribution analysis. Experiments on two benchmark datasets show that Pocket-PROTACs consistently outperforms fingerprint-based baselines and recent deep learning methods. The learned relevance maps highlight localized interaction patterns on both the POI and the E3 ligase that are qualitatively consistent with known pocket-level features. A case study on kelch domain containing 2 (KLHDC2)-engaging bromodomain and extra-terminal domain (BET)\u00a0PROTACs further demonstrates that our model accurately predicts degradation behavior and provides biologically meaningful, attention-based interpretations, offering practical support for PROTAC design and experimental investigation.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Availability and implementation<\/jats:title>\n                    <jats:p>Source code and datasets are available at https:\/\/github.com\/Adochew\/Pocket-PROTACs.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btag431","type":"journal-article","created":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T11:53:32Z","timestamp":1781870012000},"source":"Crossref","is-referenced-by-count":0,"title":["Pocket-PROTACs: an interpretable pocket-aware deep learning framework for predicting PROTAC-induced protein degradation"],"prefix":"10.1093","volume":"42","author":[{"given":"Kai","family":"Chen","sequence":"first","affiliation":[{"name":"School of Computer Science and Engineering, Central South University , Changsha, 410083,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-1949-167X","authenticated-orcid":false,"given":"Zhijian","family":"Huang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Central South University , Changsha, 410083,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yinbo","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Central South University , Changsha, 410083,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Siyuan","family":"Shen","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Central South University , Changsha, 410083,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-9266-0415","authenticated-orcid":false,"given":"Jinmiao","family":"Song","sequence":"additional","affiliation":[{"name":"Xinjiang Key Laboratory of Intelligent Computing and Smart Applications, Xinjiang University , Urumqi 830046,","place":["China"]},{"name":"School of Software, Xinjiang University , Urumqi 830046,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2869-1619","authenticated-orcid":false,"given":"Lei","family":"Deng","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Central South University , Changsha, 410083,","place":["China"]},{"name":"Xinjiang Key Laboratory of Intelligent Computing and Smart Applications, Xinjiang University , Urumqi 830046,","place":["China"]},{"name":"School of Software, Xinjiang University , Urumqi 830046,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,8,21]]},"reference":[{"key":"2026082407025734600_btag431-B1","doi-asserted-by":"crossref","first-page":"181","DOI":"10.1038\/s41573-021-00371-6","article-title":"Protac targeted protein degraders: the past is prologue","volume":"21","author":"B\u00e9k\u00e9s","year":"2022","journal-title":"Nat Rev Drug Discov"},{"key":"2026082407025734600_btag431-B2","doi-asserted-by":"crossref","first-page":"563","DOI":"10.1021\/acs.biochem.4c00577","article-title":"Journey of protac: from bench to clinical trial and beyond","volume":"64","author":"Berkley","year":"2025","journal-title":"Biochemistry"},{"key":"2026082407025734600_btag431-B3","doi-asserted-by":"crossref","first-page":"bbae654","DOI":"10.1093\/bib\/bbae654","article-title":"Et-protacs: modeling ternary complex interactions using cross-modal learning and ternary attention for accurate protac-induced degradation prediction","volume":"26","author":"Cai","year":"2024","journal-title":"Brief Bioinform"},{"key":"2026082407025734600_btag431-B4","doi-asserted-by":"crossref","first-page":"e08138","DOI":"10.1002\/advs.202508138","article-title":"Interpretable protac degradation prediction with structure-informed deep ternary attention framework","volume":"12","author":"Chen","year":"2025","journal-title":"Adv Sci (Weinh)"},{"key":"2026082407025734600_btag431-B5","doi-asserted-by":"crossref","first-page":"514","DOI":"10.1038\/nchembio.2329","article-title":"Structural basis of protac cooperative recognition for selective protein degradation","volume":"13","author":"Gadd","year":"2017","journal-title":"Nat Chem Biol"},{"key":"2026082407025734600_btag431-B6","doi-asserted-by":"crossref","first-page":"D1510","DOI":"10.1093\/nar\/gkae768","article-title":"Protac-db 3.0: an updated database of protacs with extended pharmacokinetic parameters","volume":"53","author":"Ge","year":"2025","journal-title":"Nucleic Acids Res"},{"key":"2026082407025734600_btag431-B7","doi-asserted-by":"crossref","first-page":"2158","DOI":"10.1039\/D4DD00177J","article-title":"A comprehensive review of emerging approaches in machine learning for de novo protac design","volume":"3","author":"Gharbi","year":"2024","journal-title":"Digit Discov"},{"key":"2026082407025734600_btag431-B8","doi-asserted-by":"crossref","first-page":"103395","DOI":"10.1016\/j.drudis.2022.103395","article-title":"Protac technology: a new drug design for chemical biology with many challenges in drug discovery","volume":"28","author":"Guedeney","year":"2023","journal-title":"Drug Discov Today"},{"key":"2026082407025734600_btag431-B9","doi-asserted-by":"crossref","first-page":"323","DOI":"10.1038\/s41589-022-01177-2","article-title":"Functional e3 ligase hotspots and resistance mechanisms to small-molecule degraders","volume":"19","author":"Hanzl","year":"2023","journal-title":"Nat Chem Biol"},{"key":"2026082407025734600_btag431-B10","doi-asserted-by":"crossref","first-page":"1311","DOI":"10.1007\/s11548-018-1797-4","article-title":"Ternarynet: faster deep model inference without gpus for medical 3D segmentation using sparse and binary convolutions","volume":"13","author":"Heinrich","year":"2018","journal-title":"Int J Comput Assist Radiol Surg"},{"key":"2026082407025734600_btag431-B11","first-page":"32","article-title":"Cross attention network for few-shot classification","author":"Hou","year":"2019","journal-title":"Adv Neural Inf Process Syst"},{"key":"2026082407025734600_btag431-B12","doi-asserted-by":"crossref","first-page":"9805","DOI":"10.1038\/s41467-025-64773-5","article-title":"Protac repurposing uncovers a noncanonical binding surface that mediates chemical degradation of nuclear receptors","volume":"16","author":"Huber","year":"2025","journal-title":"Nat Commun"},{"key":"2026082407025734600_btag431-B13","doi-asserted-by":"crossref","first-page":"484","DOI":"10.1177\/2472555220965528","article-title":"E3 ligase ligands for protacs: how they were found and how to discover new ones","volume":"26","author":"Ishida","year":"2021","journal-title":"SLAS Discov"},{"key":"2026082407025734600_btag431-B14","doi-asserted-by":"crossref","first-page":"39","DOI":"10.1186\/s13321-018-0285-8","article-title":"P2rank: machine learning based tool for rapid and accurate prediction of ligand binding sites from protein structure","volume":"10","author":"Kriv\u00e1k","year":"2018","journal-title":"J Cheminform"},{"key":"2026082407025734600_btag431-B15","doi-asserted-by":"crossref","first-page":"7133","DOI":"10.1038\/s41467-022-34807-3","article-title":"Deepprotacs is a deep learning-based targeted degradation predictor for protacs","volume":"13","author":"Li","year":"2022","journal-title":"Nat Commun"},{"key":"2026082407025734600_btag431-B16","doi-asserted-by":"crossref","first-page":"308","DOI":"10.1016\/j.cels.2020.03.002","article-title":"Monn: a multi-objective neural network for predicting compound-protein interactions and affinities","volume":"10","author":"Li","year":"2020","journal-title":"Cell Syst"},{"key":"2026082407025734600_btag431-B17","doi-asserted-by":"crossref","first-page":"e2401051","DOI":"10.1002\/marc.202401051","article-title":"Integrating proteolysis-targeting chimeras (protacs) with delivery systems for more efficient and precise targeted protein degradation","volume":"46","author":"Lin","year":"2025","journal-title":"Macromol Rapid Commun"},{"key":"2026082407025734600_btag431-B18","doi-asserted-by":"crossref","first-page":"110040","DOI":"10.1016\/j.asoc.2023.110040","article-title":"EGNN: graph structure learning based on evolutionary computation helps more in graph neural networks","volume":"135","author":"Liu","year":"2023","journal-title":"Appl Soft Comput"},{"key":"2026082407025734600_btag431-B19","doi-asserted-by":"crossref","first-page":"8595","DOI":"10.1038\/s41467-025-63713-7","article-title":"Frustration in the protein-protein interface plays a Central role in the cooperativity of protac ternary complexes","volume":"16","author":"Ma","year":"2025","journal-title":"Nat Commun"},{"key":"2026082407025734600_btag431-B20","doi-asserted-by":"crossref","first-page":"878762","DOI":"10.3389\/fimmu.2022.878762","article-title":"Pandora: a fast, anchor-restrained modelling protocol for peptide: mhc complexes","volume":"13","author":"Marzella","year":"2022","journal-title":"Front Immunol"},{"key":"2026082407025734600_btag431-B21","doi-asserted-by":"crossref","first-page":"138","DOI":"10.1016\/j.pharmthera.2017.02.027","article-title":"Targeted protein degradation by protacs","volume":"174","author":"Neklesa","year":"2017","journal-title":"Pharmacol Ther"},{"key":"2026082407025734600_btag431-B22","doi-asserted-by":"crossref","first-page":"7971","DOI":"10.1039\/D2CS00478J","article-title":"Protac degraders as chemical probes for studying target biology and target validation","volume":"51","author":"N\u011bmec","year":"2022","journal-title":"Chem Soc Rev"},{"key":"2026082407025734600_btag431-B23","doi-asserted-by":"crossref","first-page":"1793","DOI":"10.3390\/ph18121793","article-title":"Advancing protac discovery through artificial intelligence: opportunities, challenges, and future directions","volume":"18","author":"Park","year":"2025","journal-title":"Pharmaceuticals"},{"key":"2026082407025734600_btag431-B24","doi-asserted-by":"crossref","first-page":"1350","DOI":"10.1074\/mcp.T300009-MCP200","article-title":"Development of protacs to target cancer-promoting proteins for ubiquitination and degradation","volume":"2","author":"Sakamoto","year":"2003","journal-title":"Mol Cell Proteomics"},{"key":"2026082407025734600_btag431-B25","doi-asserted-by":"crossref","first-page":"8829","DOI":"10.1038\/s41467-024-52966-3","article-title":"Principles of paralog-specific targeted protein degradation engaging the c-degron e3 klhdc2","volume":"15","author":"Scott","year":"2024","journal-title":"Nat Commun"},{"key":"2026082407025734600_btag431-B26","doi-asserted-by":"crossref","first-page":"273","DOI":"10.37349\/etat.2020.00018","article-title":"Current strategies for the design of protac linkers: a critical review","volume":"1","author":"Troup","year":"2020","journal-title":"Explor Target Antitumor Ther"},{"key":"2026082407025734600_btag431-B27","doi-asserted-by":"crossref","first-page":"218","DOI":"10.1038\/s42004-025-01598-9","article-title":"Synthesis, biological evaluation and clinical trials of cereblon-based protacs","volume":"8","author":"Vicente","year":"2025","journal-title":"Commun Chem"},{"key":"2026082407025734600_btag431-B28","doi-asserted-by":"crossref","first-page":"e202300163","DOI":"10.1002\/cbic.202300163","article-title":"Biophysical and computational approaches to study ternary complexes: a \u2018cooperative relationship\u2019to rationalize targeted protein degradation","volume":"24","author":"Ward","year":"2023","journal-title":"Chembiochem"},{"key":"2026082407025734600_btag431-B29","doi-asserted-by":"crossref","first-page":"D1381","DOI":"10.1093\/nar\/gkaa807","article-title":"Protac-db: an online database of protacs","volume":"49","author":"Weng","year":"2021","journal-title":"Nucleic Acids Res"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/42\/Supplement_2\/btag431\/70726200\/btag431.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/42\/Supplement_2\/btag431\/70726200\/btag431.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,24]],"date-time":"2026-08-24T11:03:07Z","timestamp":1787569387000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/doi\/10.1093\/bioinformatics\/btag431\/8767291"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,1]]},"references-count":29,"journal-issue":{"issue":"Supplement_2","published-print":{"date-parts":[[2026,8,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btag431","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,8]]},"published":{"date-parts":[[2026,8,1]]},"article-number":"btag431"}}