{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T15:55:05Z","timestamp":1787327705476,"version":"build-2736575974"},"reference-count":47,"publisher":"Wiley","license":[{"start":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T00:00:00Z","timestamp":1787270400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T00:00:00Z","timestamp":1787270400000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/501100006469","name":"Fundo para o Desenvolvimento das Ci\u00eancias e da Tecnologia","doi-asserted-by":"publisher","award":["0043\/2023\/AFJ"],"award-info":[{"award-number":["0043\/2023\/AFJ"]}],"id":[{"id":"10.13039\/501100006469","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["advanced.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Advanced Science"],"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:p>\n                    Accurately predicting drug\u2013target affinity (DTA) is crucial for accelerating virtual screening and guiding lead optimization in drug discovery. However, current computational approaches face a critical trade\u2010off: interaction\u2010free models lack fine\u2010grained binding details, while interaction\u2010based models overlook higher\u2010order contextual and functional patterns. This limitation hinders both prediction performance and real\u2010world generalization. To overcome this, we propose MF\u2010Net, a unified hierarchical multiscale fusion framework that integrates sequence\u2010, atomic\u2010, and fragment\u2010level representations to model drug\u2013target interactions across complementary scales. MF\u2010Net achieves state\u2010of\u2010the\u2010art performance on the PDBBind v2016 benchmark and demonstrates strong early enrichment across multiple virtual screening datasets. Additionally, ADP\u2010Glo assays confirm that the MF\u2010Net\u2010guided virtual screening pipeline identifies seven novel nanomolar inhibitors targeting hematopoietic progenitor kinase 1 (HPK1). Among them, one compound achieves sub\u2010nanomolar activity (IC\n                    <jats:sub>50<\/jats:sub>\n                    = 0.41\u00a0nM), outperforming the positive control inhibitor Sunitinib. These results demonstrate that MF\u2010Net not only excels on standard benchmarks but also delivers tangible lead discovery outcomes, underscoring its practical value for structure\u2010based drug design.\n                  <\/jats:p>","DOI":"10.1002\/advs.77345","type":"journal-article","created":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T14:50:19Z","timestamp":1787323819000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["A Unified Hierarchical Multiscale Fusion Framework for Drug\u2013Target Affinity Prediction: From Benchmark Performance to Nanomolar Inhibitor Discovery"],"prefix":"10.1002","author":[{"given":"Shuo","family":"Liu","sequence":"first","affiliation":[{"name":"Chinese Medicine Guangdong Laboratory (Hengqin Laboratory) Guangdong\u2010Macao In\u2010Depth Cooperation Zone in Hengqin  Guangdong China"},{"name":"The Second Affiliated Hospital of Guangzhou University of Chinese Medicine  Guangzhou China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-0967-5262","authenticated-orcid":false,"given":"Xiang","family":"Zhang","sequence":"additional","affiliation":[{"name":"Chinese Medicine Guangdong Laboratory (Hengqin Laboratory) Guangdong\u2010Macao In\u2010Depth Cooperation Zone in Hengqin  Guangdong China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Haixia","family":"Feng","sequence":"additional","affiliation":[{"name":"Chinese Medicine Guangdong Laboratory (Hengqin Laboratory) Guangdong\u2010Macao In\u2010Depth Cooperation Zone in Hengqin  Guangdong China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2756-0449","authenticated-orcid":false,"given":"Yuquan","family":"Li","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Public Big Data College of Computer Science and Technology Guizhou University  Guiyang China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaoqing","family":"Gong","sequence":"additional","affiliation":[{"name":"Faculty of Applied Sciences Macao Polytechnic University  Macao China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yong","family":"Liang","sequence":"additional","affiliation":[{"name":"Chinese Medicine Guangdong Laboratory (Hengqin Laboratory) Guangdong\u2010Macao In\u2010Depth Cooperation Zone in Hengqin  Guangdong China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8974-0173","authenticated-orcid":false,"given":"Xiaojun","family":"Yao","sequence":"additional","affiliation":[{"name":"Faculty of Applied Sciences Macao Polytechnic University  Macao China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9284-3667","authenticated-orcid":false,"given":"Huanxiang","family":"Liu","sequence":"additional","affiliation":[{"name":"Faculty of Applied Sciences Macao Polytechnic University  Macao China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,8,21]]},"reference":[{"key":"e_1_2_9_2_1","first-page":"1","article-title":"Sequence\u2010Based Virtual Screening Using Transformers","volume":"16","author":"Zhang S.","year":"2025","journal-title":"Nature Communications"},{"key":"e_1_2_9_3_1","first-page":"1","article-title":"A Multi\u2010Task Self\u2010Supervised Strategy for Predicting Molecular Properties and FGFR1 Inhibitors","volume":"12","author":"Yang X.","year":"2025","journal-title":"Advancement of Science"},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41467\u2010025\u201059917\u20106"},{"key":"e_1_2_9_5_1","doi-asserted-by":"publisher","DOI":"10.1126\/sciadv.abo5128"},{"key":"e_1_2_9_6_1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.4c01528"},{"key":"e_1_2_9_7_1","doi-asserted-by":"publisher","DOI":"10.1186\/s12859\u2010025\u201006165\u20106"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41467\u2010024\u201054440\u20106"},{"key":"e_1_2_9_9_1","doi-asserted-by":"publisher","DOI":"10.1021\/acsomega.4c08048"},{"key":"e_1_2_9_10_1","first-page":"5350","article-title":"Multi\u2010Modal Deep Representation Learning Accurately Identifies and Interprets Drug\u2013Target Interactions","volume":"29","author":"Hu J.","year":"2025","journal-title":"IEEE J Biomed Heal Informatics"},{"key":"e_1_2_9_11_1","doi-asserted-by":"publisher","DOI":"10.1186\/s12864\u2010025\u201011234\u20104"},{"key":"e_1_2_9_12_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41589\u2010025\u201001929\u2010w"},{"key":"e_1_2_9_13_1","doi-asserted-by":"publisher","DOI":"10.1021\/acscentsci.3c00572"},{"key":"e_1_2_9_14_1","doi-asserted-by":"publisher","DOI":"10.1002\/advs.202405404"},{"key":"e_1_2_9_15_1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jctc.6b00991"},{"key":"e_1_2_9_16_1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jmedchem.2c00460"},{"key":"e_1_2_9_17_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256\u2010023\u201000751\u20100"},{"key":"e_1_2_9_18_1","doi-asserted-by":"publisher","DOI":"10.1039\/D0RA02297G"},{"key":"e_1_2_9_19_1","doi-asserted-by":"publisher","DOI":"10.1038\/s42256\u2010022\u201000605\u20101"},{"key":"e_1_2_9_20_1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.4c00403"},{"key":"e_1_2_9_21_1","doi-asserted-by":"publisher","DOI":"10.1039\/D1SC05180F"},{"key":"e_1_2_9_22_1","doi-asserted-by":"crossref","first-page":"i821","DOI":"10.1093\/bioinformatics\/bty593","article-title":"Deep Drug\u2013Target Binding Affinity Prediction","volume":"34","author":"\u00d6zt\u00fcrk H.","year":"2018","journal-title":"Bioinformatics"},{"key":"e_1_2_9_23_1","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btaa921"},{"key":"e_1_2_9_24_1","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btaa982"},{"key":"e_1_2_9_25_1","doi-asserted-by":"publisher","DOI":"10.1039\/D1SC06946B"},{"issue":"12","key":"e_1_2_9_26_1","doi-asserted-by":"crossref","first-page":"8191","DOI":"10.1109\/TPAMI.2024.3400515","article-title":"Interaction\u2010Based Inductive Bias in Graph Neural Networks: Enhancing Protein\u2013Ligand Binding Affinity Predictions from 3D Structures","volume":"46","author":"Yang Z.","year":"2024","journal-title":"Trans Pattern Anal Mach Intell"},{"key":"e_1_2_9_27_1","doi-asserted-by":"publisher","DOI":"10.1021\/acsomega.9b01997"},{"key":"e_1_2_9_28_1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jmedchem.1c01830"},{"key":"e_1_2_9_29_1","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/bty374"},{"key":"e_1_2_9_30_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2026.105061"},{"key":"e_1_2_9_31_1","doi-asserted-by":"crossref","first-page":"459","DOI":"10.1186\/s12859-022-04812-w","article-title":"A Novel Method for Drug\u2013Target Interaction Prediction Based on Graph Transformers Model","volume":"23","author":"Wang H.","year":"2022","journal-title":"BMC Bioinformatics"},{"key":"e_1_2_9_32_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13321\u2010024\u201000938\u20106"},{"key":"e_1_2_9_33_1","doi-asserted-by":"publisher","DOI":"10.3390\/ijms24032649"},{"key":"e_1_2_9_34_1","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbad462"},{"key":"e_1_2_9_35_1","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btae413"},{"key":"e_1_2_9_36_1","first-page":"1","article-title":"Predicting or Pretending: Artificial Intelligence for Protein\u2013Ligand Interactions Lack of Sufficiently Large and Unbiased Datasets","volume":"11","author":"Yang J.","year":"2020","journal-title":"Frontiers in Pharmacology"},{"key":"e_1_2_9_37_1","first-page":"1","article-title":"Hidden Bias in the DUD\u2010E Dataset Leads to Misleading Performance of Deep Learning in Structure\u2010Based Virtual Screening","volume":"14","author":"Chen L.","year":"2019","journal-title":"PLoS ONE"},{"key":"e_1_2_9_38_1","doi-asserted-by":"crossref","first-page":"2788","DOI":"10.1021\/acs.jcim.1c00292","article-title":"True Accuracy of Fast Scoring Functions to Predict High\u2010Throughput Screening Data From Docking Poses: The Simpler the Better","volume":"61","author":"VK T.\u2010N.","year":"2021","journal-title":"Journal of Chemical Information and Modeling"},{"key":"e_1_2_9_39_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41467\u2010025\u201065869\u20108"},{"key":"e_1_2_9_40_1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.8b00545"},{"key":"e_1_2_9_41_1","doi-asserted-by":"publisher","DOI":"10.1021\/jm048957q"},{"key":"e_1_2_9_42_1","first-page":"235","article-title":"The Protein Data Bank","volume":"28","author":"Saur I. M. L.","year":"2000","journal-title":"Nature Reviews Immunology"},{"key":"e_1_2_9_43_1","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbae145","article-title":"A New Paradigm for Applying Deep Learning to Protein\u2013Ligand Interaction Prediction","volume":"25","author":"Wang Z.","year":"2024","journal-title":"Brief Bioinform"},{"key":"e_1_2_9_44_1","doi-asserted-by":"publisher","DOI":"10.1021\/jm300687e"},{"key":"e_1_2_9_45_1","doi-asserted-by":"publisher","DOI":"10.1021\/ci400115b"},{"key":"e_1_2_9_46_1","doi-asserted-by":"crossref","first-page":"4263","DOI":"10.1021\/acs.jcim.0c00155","article-title":"LIT\u2010PCBA: An Unbiased Data Set for Machine Learning and Virtual Screening","volume":"60","author":"VK T.\u2010N.","year":"2020","journal-title":"Journal of Chemical Information and Modeling"},{"key":"e_1_2_9_47_1","doi-asserted-by":"publisher","DOI":"10.1021\/acs.jcim.7b00403"},{"key":"e_1_2_9_48_1","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbaa410"}],"container-title":["Advanced Science"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/advanced.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/advs.77345","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/advanced.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/advs.77345","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/advanced.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/advs.77345","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T14:50:23Z","timestamp":1787323823000},"score":1,"resource":{"primary":{"URL":"https:\/\/advanced.onlinelibrary.wiley.com\/doi\/10.1002\/advs.77345"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,21]]},"references-count":47,"alternative-id":["10.1002\/advs.77345"],"URL":"https:\/\/doi.org\/10.1002\/advs.77345","archive":["Portico"],"relation":{},"ISSN":["2198-3844","2198-3844"],"issn-type":[{"value":"2198-3844","type":"print"},{"value":"2198-3844","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,21]]},"assertion":[{"value":"2026-05-08","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-07-22","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-08-21","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e77345"}}