{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T12:58:55Z","timestamp":1791464335599,"version":"4.3.4"},"reference-count":28,"publisher":"Oxford University Press (OUP)","issue":"5","license":[{"start":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T00:00:00Z","timestamp":1791417600000},"content-version":"vor","delay-in-days":37,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Postgraduate Research & Practice Innovation Program of Jiangsu Province","award":["KYCX25_1091"],"award-info":[{"award-number":["KYCX25_1091"]}]},{"name":"Research on Key Technologies for Monitoring and Identifying Drug Abuse of Anesthetic Drugs and Psychotropic Drugs, and Intervention for Addiction","award":["2023YFC3304200"],"award-info":[{"award-number":["2023YFC3304200"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,9,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Peptide drug repurposing depends on identifying likely protein targets for a query peptide across a large candidate space, yet most existing peptide\u2013protein interaction models are optimized for isolated pairwise classification rather than target prioritization. Here, we present PEARL, a recommendation-oriented heterogeneous graph learning framework for peptide target recommendation. PEARL represents peptides and proteins as distinct node types connected by known peptide-protein associations and multi-view peptide similarity, and combines graph-based representation learning with similarity-aware score refinement to rank candidate targets for a query peptide. On the UMPPI benchmark, PEARL achieved mean area under the receiver operating characteristic curve (AUC), area under the precision\u2013recall curve (AUPR), and Matthews correlation coefficient values of 0.868, 0.855, and 0.571 in five-fold cross-validation, outperforming the strongest baseline, HGT-PepPI, by 4.4, 4.8, and 7.0 percentage points, respectively. Under clustered cold-start evaluation, PEARL delivered the best overall AUC across the novel-protein, novel-peptide, and novel-pair settings, indicating robust generalization under increasing sequence dissimilarity. PEARL also remained the top performer on two independent test sets, reaching an AUC of 0.863 on Test167 and 0.871 on Test1440. Additional analyses showed that PEARL can be effectively calibrated and that its learned latent space groups functionally related targets beyond simple sequence identity. In a case study of the unseen therapeutic peptide CIGB-814, PEARL ranked SPSB2 as the top candidate target. Docking and molecular dynamics simulations supported a stable binding mode, and surface plasmon resonance further confirmed direct binding with an apparent KD of 5.538\u00a0\u03bcM. Together, these results support PEARL as a scalable computational framework for peptide target recommendation and peptide drug repurposing.<\/jats:p>","DOI":"10.1093\/bib\/bbag551","type":"journal-article","created":{"date-parts":[[2026,9,27]],"date-time":"2026-09-27T11:31:35Z","timestamp":1790508695000},"source":"Crossref","is-referenced-by-count":0,"title":["PEARL: novel protein targets recommendation model via heterogeneous graph representation learning"],"prefix":"10.1093","volume":"27","author":[{"given":"Zixuan","family":"Liu","sequence":"first","affiliation":[{"name":"School of Artificial Intelligence, China Pharmaceutical University , 639 Longmian Avenue, Jiangning District, Nanjing 211198, Jiangsu ,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiasi","family":"Luan","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence, China Pharmaceutical University , 639 Longmian Avenue, Jiangning District, Nanjing 211198, Jiangsu ,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhen","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence, China Pharmaceutical University , 639 Longmian Avenue, Jiangning District, Nanjing 211198, Jiangsu ,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kui","family":"Tang","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence, China Pharmaceutical University , 639 Longmian Avenue, Jiangning District, Nanjing 211198, Jiangsu ,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xingyu","family":"Qin","sequence":"additional","affiliation":[{"name":"New Drug Screening and Pharmacodynamics Evaluation Center, State Key Laboratory of Natural Medicines, China Pharmaceutical University , 639 Longmian Avenue, Jiangning District, Nanjing 210009 ,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhixiao","family":"Qi","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence, China Pharmaceutical University , 639 Longmian Avenue, Jiangning District, Nanjing 211198, Jiangsu ,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Borui","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence, China Pharmaceutical University , 639 Longmian Avenue, Jiangning District, Nanjing 211198, Jiangsu ,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0617-5840","authenticated-orcid":false,"given":"Jun","family":"Liao","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence, China Pharmaceutical University , 639 Longmian Avenue, Jiangning District, Nanjing 211198, Jiangsu ,","place":["China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,10,8]]},"reference":[{"key":"2026100808260261900_ref1","doi-asserted-by":"publisher","first-page":"6815","DOI":"10.3390\/ijms26146815","article-title":"Innovative peptide therapeutics in the pipeline: transforming cancer detection and treatment","volume":"26","author":"Nsereko","year":"2025","journal-title":"Int J Mol Sci"},{"key":"2026100808260261900_ref2","doi-asserted-by":"publisher","first-page":"309","DOI":"10.1038\/s41573-020-00135-8","article-title":"Trends in peptide drug discovery","volume":"20","author":"Muttenthaler","year":"2021","journal-title":"Nat Rev Drug Discov"},{"key":"2026100808260261900_ref3","doi-asserted-by":"publisher","first-page":"122","DOI":"10.1016\/j.drudis.2014.10.003","article-title":"Peptide therapeutics: current status and future directions","volume":"20","author":"Fosgerau","year":"2015","journal-title":"Drug Discov Today"},{"key":"2026100808260261900_ref4","doi-asserted-by":"publisher","first-page":"430","DOI":"10.3390\/molecules26020430","article-title":"A global review on short peptides: frontiers and perspectives","volume":"26","author":"Apostolopoulos","year":"2021","journal-title":"Molecules"},{"key":"2026100808260261900_ref5","doi-asserted-by":"publisher","first-page":"1293","DOI":"10.1016\/S1875-5364(25)60945-9","article-title":"MolP-PC: a multi-view fusion and multi-task learning framework for drug ADMET property prediction","volume":"23","author":"Li","year":"2025","journal-title":"Chin J Nat Med"},{"key":"2026100808260261900_ref6","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1038\/nrd.2018.168","article-title":"Drug repurposing: progress, challenges and recommendations","volume":"18","author":"Pushpakom","year":"2019","journal-title":"Nat Rev Drug Discov"},{"key":"2026100808260261900_ref7","doi-asserted-by":"publisher","first-page":"2076","DOI":"10.1016\/j.drudis.2019.06.014","article-title":"Drug repurposing: a promising tool to accelerate the drug discovery process","volume":"24","author":"Parvathaneni","year":"2019","journal-title":"Drug Discov Today"},{"key":"2026100808260261900_ref8","doi-asserted-by":"publisher","first-page":"894","DOI":"10.1016\/j.sbi.2013.07.006","article-title":"Peptide docking and structure-based characterization of peptide binding: from knowledge to know-how","volume":"23","author":"London","year":"2013","journal-title":"Curr Opin Struct Biol"},{"key":"2026100808260261900_ref9","doi-asserted-by":"publisher","first-page":"bbae297","DOI":"10.1093\/bib\/bbae297","article-title":"GAPS: a geometric attention-based network for peptide binding site identification by the transfer learning approach","volume":"25","author":"Zhu","year":"2024","journal-title":"Brief Bioinform"},{"key":"2026100808260261900_ref10","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btae708","article-title":"TPepPro: a deep learning model for predicting peptide\u2013protein interactions","volume":"41","author":"Jin","year":"2024","journal-title":"Bioinformatics"},{"key":"2026100808260261900_ref11","doi-asserted-by":"publisher","first-page":"1530","DOI":"10.1016\/j.drudis.2018.05.006","article-title":"Protein-peptide docking: opportunities and challenges","volume":"23","author":"Ciemny","year":"2018","journal-title":"Drug Discov Today"},{"key":"2026100808260261900_ref12","doi-asserted-by":"publisher","first-page":"401","DOI":"10.1039\/d3cb00208j","article-title":"Leveraging machine learning models for peptide-protein interaction prediction","volume":"5","author":"Yin","year":"2024","journal-title":"RSC Chem Biol"},{"key":"2026100808260261900_ref13","doi-asserted-by":"publisher","first-page":"763102","DOI":"10.3389\/fbinf.2021.763102","article-title":"InterPepRank: assessment of docked peptide conformations by a deep graph network","volume":"1","author":"Johansson-Akhe","year":"2021","journal-title":"Front Bioinform"},{"key":"2026100808260261900_ref14","doi-asserted-by":"publisher","first-page":"370","DOI":"10.1186\/s12859-022-04910-9","article-title":"Struct2Graph: a graph attention network for structure based predictions of protein-protein interactions","volume":"23","author":"Baranwal","year":"2022","journal-title":"BMC Bioinformatics"},{"key":"2026100808260261900_ref15","doi-asserted-by":"publisher","first-page":"5465","DOI":"10.1038\/s41467-021-25772-4","article-title":"A deep-learning framework for multi-level peptide-protein interaction prediction","volume":"12","author":"Lei","year":"2021","journal-title":"Nat Commun"},{"key":"2026100808260261900_ref16","doi-asserted-by":"publisher","first-page":"503","DOI":"10.1038\/s42003-022-03445-2","article-title":"PepNN: a deep attention model for the identification of peptide binding sites","volume":"5","author":"Abdin","year":"2022","journal-title":"Commun Biol"},{"key":"2026100808260261900_ref17","doi-asserted-by":"publisher","first-page":"e2216697120","DOI":"10.1073\/pnas.2216697120","article-title":"Peptide-binding specificity prediction using fine-tuned protein structure prediction networks","volume":"120","author":"Motmaen","year":"2023","journal-title":"Proc Natl Acad Sci U S A"},{"key":"2026100808260261900_ref18","doi-asserted-by":"publisher","first-page":"3789","DOI":"10.1021\/acs.jcim.4c02365","article-title":"UMPPI: unveiling multilevel protein-peptide interaction prediction via language models","volume":"65","author":"Xiong","year":"2025","journal-title":"J Chem Inf Model"},{"key":"2026100808260261900_ref19","doi-asserted-by":"publisher","first-page":"btae290","DOI":"10.1093\/bioinformatics\/btae290","article-title":"LMCrot: an enhanced protein crotonylation site predictor by leveraging an interpretable window-level embedding from a transformer-based protein language model","volume":"40","author":"Pratyush","year":"2024","journal-title":"Bioinformatics"},{"key":"2026100808260261900_ref20","doi-asserted-by":"publisher","first-page":"1009","DOI":"10.1109\/TPDS.2012.194","article-title":"Retrieving smith-waterman alignments with optimizations for megabase biological sequences using GPU","volume":"24","author":"O","year":"2012","journal-title":"IEEE Trans Parallel Distrib Syst"},{"key":"2026100808260261900_ref21","doi-asserted-by":"publisher","first-page":"3479","DOI":"10.1109\/TCAD.2025.3543436","article-title":"A lightweight heterogeneous graph embedding framework for hotspot detection","volume":"44","author":"Yan","year":"2025","journal-title":"IEEE Trans Comput Aided Des Integr Circuits Syst"},{"key":"2026100808260261900_ref22","doi-asserted-by":"publisher","first-page":"10727","DOI":"10.1007\/s00500-024-09797-7","article-title":"Subgraph generation applied in GraphSAGE deal with imbalanced node classification","volume":"28","author":"Huang","year":"2024","journal-title":"Soft Comput"},{"key":"2026100808260261900_ref23","doi-asserted-by":"publisher","first-page":"883","DOI":"10.1038\/nprot.2010.32","article-title":"The HADDOCK web server for data-driven biomolecular docking","volume":"5","author":"Vries","year":"2010","journal-title":"Nat Protoc"},{"key":"2026100808260261900_ref24","doi-asserted-by":"publisher","first-page":"1157","DOI":"10.1126\/science.163.3872.1157","article-title":"Amber: a botanical inquiry: Amber provides an evolutionary framework for interdisciplinary studies of resin-secreting plants","volume":"163","author":"Langenheim","year":"1969","journal-title":"Science"},{"key":"2026100808260261900_ref25","doi-asserted-by":"publisher","first-page":"1630","DOI":"10.2174\/156802607782341073","article-title":"SPR-based fragment screening: advantages and applications","volume":"7","author":"Neumann","year":"2007","journal-title":"Curr Top Med Chem"},{"key":"2026100808260261900_ref26","doi-asserted-by":"publisher","first-page":"471","DOI":"10.3109\/08916934.2010.550590","article-title":"An altered peptide ligand corresponding to a novel epitope from heat-shock protein 60 induces regulatory T cells and suppresses pathogenic response in an animal model of adjuvant-induced arthritis","volume":"44","author":"Dominguez Mdel","year":"2011","journal-title":"Autoimmunity"},{"key":"2026100808260261900_ref27","doi-asserted-by":"publisher","first-page":"130","DOI":"10.1016\/j.jpba.2017.05.030","article-title":"Development and validation of a bioanalytical method based on LC-MS\/MS analysis for the quantitation of CIGB-814 peptide in plasma from rheumatoid arthritis patients","volume":"143","author":"Cabrales-Rico","year":"2017","journal-title":"J Pharm Biomed Anal"},{"key":"2026100808260261900_ref28","doi-asserted-by":"publisher","first-page":"955","DOI":"10.1007\/s10067-018-4360-3","article-title":"CIGB-814, an altered peptide ligand derived from human heat-shock protein 60, decreases anti-cyclic citrullinated peptides antibodies in patients with rheumatoid arthritis","volume":"38","author":"Corrales","year":"2019","journal-title":"Clin Rheumatol"}],"container-title":["Briefings in Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/27\/5\/bbag551\/71717960\/bbag551.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/27\/5\/bbag551\/71717960\/bbag551.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T12:26:07Z","timestamp":1791462367000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bib\/article\/doi\/10.1093\/bib\/bbag551\/8885513"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,1]]},"references-count":28,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2026,9,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bib\/bbag551","relation":{},"ISSN":["1467-5463","1477-4054"],"issn-type":[{"value":"1467-5463","type":"print"},{"value":"1477-4054","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,9]]},"published":{"date-parts":[[2026,9,1]]},"article-number":"bbag551"}}