{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T12:02:50Z","timestamp":1791460970365,"version":"4.3.4"},"reference-count":41,"publisher":"Wiley","license":[{"start":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T00:00:00Z","timestamp":1791417600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"},{"start":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T00:00:00Z","timestamp":1791417600000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62373216"],"award-info":[{"award-number":["62373216"]}],"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":["92374107"],"award-info":[{"award-number":["92374107"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100012166","name":"National Key Research and Development Program of China","doi-asserted-by":"publisher","award":["2022YFA1004801"],"award-info":[{"award-number":["2022YFA1004801"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Proteins"],"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:p>\n                    Accurate prediction of nucleic acid\u2010binding sites is essential for understanding protein function and facilitating drug discovery. However, current computational methods are limited by single\u2010layer language model representations, C\u03b1 point clouds that discard local\u2010frame geometry, and feature fusion in which high\u2010dimensional sequence embeddings overshadow weaker evolutionary and structural signals. Here, we present PNBind, a multimodal deep learning framework that leverages protein language models, evolutionary information, and three\u2010dimensional structural features via a geometric vector perceptron (GVP) backbone with an invariant point attention (IPA) layer for nucleic acid\u2010binding site prediction. By integrating sequence representations from the ESM\u2010series language models with alignment\u2010derived evolutionary features, secondary\u2010structure data, and geometric features extracted from five\u2010atom local frames, PNBind enables effective late\u2010stage multimodal signal fusion. Comprehensive benchmarking on the DNA\u2010129, DNA\u2010181, RNA\u2010117, and RNA\u2010285 datasets reveals that PNBind achieves strong performance, with\n                    <jats:italic>F<\/jats:italic>\n                    1 scores of 0.611, 0.443, 0.370, and 0.472 and MCC values of 0.586, 0.425, 0.351, and 0.416, respectively. Ablation studies confirm the dominant contribution of the protein language model backbone, while systematic evaluation of ESM\u2010family protein language models demonstrates that proper PLM selection significantly boosts prediction accuracy. This study demonstrates that the effective integration of protein language models and geometric deep learning offers a practical approach for accurate nucleic acid\u2010binding site prediction.\n                  <\/jats:p>","DOI":"10.1002\/prot.70190","type":"journal-article","created":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T11:46:21Z","timestamp":1791459981000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["<scp>PNBind<\/scp>\n                    : Prediction of Nucleic Acid\u2010Binding Sites Using Protein Structure and Protein Language Models"],"prefix":"10.1002","author":[{"given":"Yunhai","family":"Li","sequence":"first","affiliation":[{"name":"Department of Biomedical Engineering School of Control Science and Engineering, Shandong University  Jinan Shandong China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Guanghong","family":"Dang","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering School of Control Science and Engineering, Shandong University  Jinan Shandong China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bowen","family":"Shao","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering School of Control Science and Engineering, Shandong University  Jinan Shandong China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pengpai","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering School of Control Science and Engineering, Shandong University  Jinan Shandong China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7742-9161","authenticated-orcid":false,"given":"Zhi\u2010Ping","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Biomedical Engineering School of Control Science and Engineering, Shandong University  Jinan Shandong China"},{"name":"National Center for Applied Mathematics, Shandong University  Jinan Shandong China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,10,8]]},"reference":[{"key":"e_1_2_8_2_1","volume-title":"Essential Cell Biology","author":"Alberts B.","year":"2015"},{"key":"e_1_2_8_3_1","volume-title":"Proteins: Structure and Function","author":"Whitford D.","year":"2013"},{"key":"e_1_2_8_4_1","doi-asserted-by":"crossref","first-page":"113","DOI":"10.1007\/s40484-018-0139-4","article-title":"Towards Precise Reconstruction of Gene Regulatory Networks by Data Integration","volume":"6","author":"Liu Z.\u2010P.","year":"2018","journal-title":"Quantitative Biology"},{"key":"e_1_2_8_5_1","doi-asserted-by":"crossref","first-page":"831","DOI":"10.2174\/1574893615666200224095925","article-title":"Predicting lncRNA\u2013Protein Interactions by Machine Learning Methods: A Review","volume":"15","author":"Liu Z.\u2010P.","year":"2020","journal-title":"Current Bioinformatics"},{"key":"e_1_2_8_6_1","doi-asserted-by":"publisher","DOI":"10.1007\/978\u20101\u20100716\u20104290\u20100_16"},{"key":"e_1_2_8_7_1","doi-asserted-by":"crossref","first-page":"604","DOI":"10.2174\/1568026615666150819110703","article-title":"Prediction and Dissection of Protein\u2013RNA Interactions by Molecular Descriptors","volume":"16","author":"Liu Z.\u2010P.","year":"2016","journal-title":"Current Topics in Medicinal Chemistry"},{"key":"e_1_2_8_8_1","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pone.0097725","article-title":"RNABindRPlus: A Predictor That Combines Machine Learning and Sequence Homology\u2010Based Methods to Improve Reliability of Predicted RNA\u2010Binding Residues in Proteins","volume":"9","author":"Walia R. R.","year":"2014","journal-title":"PLoS One"},{"key":"e_1_2_8_9_1","doi-asserted-by":"crossref","first-page":"3057","DOI":"10.1021\/acs.jcim.8b00749","article-title":"DNAPred: Accurate Identification of DNA\u2010Binding Sites From Protein Sequence by Ensembled Hyperplane\u2010Distance\u2010Based Support Vector Machines","volume":"59","author":"Zhu Y. H.","year":"2019","journal-title":"Journal of Chemical Information and Modeling"},{"key":"e_1_2_8_10_1","doi-asserted-by":"crossref","first-page":"568","DOI":"10.1038\/s42003-025-07534-w","article-title":"TransBind Allows Precise Detection of DNA\u2010Binding Proteins and Residues Using Language Models and Deep Learning","volume":"8","author":"Tahmid M. T.","year":"2025","journal-title":"Communications Biology"},{"key":"e_1_2_8_11_1","doi-asserted-by":"crossref","first-page":"1885","DOI":"10.1002\/prot.24330","article-title":"DNABind: A Hybrid Algorithm for Structure\u2010Based Prediction of DNA\u2010Binding Residues by Combining Machine Learning\u2010 and Template\u2010Based Approaches","volume":"81","author":"Liu R.","year":"2013","journal-title":"Proteins"},{"key":"e_1_2_8_12_1","doi-asserted-by":"crossref","first-page":"W438","DOI":"10.1093\/nar\/gky439","article-title":"COACH\u2010D: Improved Protein\u2013Ligand Binding Sites Prediction With Refined Ligand\u2010Binding Poses Through Molecular Docking","volume":"46","author":"Wu Q.","year":"2018","journal-title":"Nucleic Acids Research"},{"key":"e_1_2_8_13_1","doi-asserted-by":"crossref","first-page":"297","DOI":"10.1007\/s12551-024-01201-w","article-title":"Structure\u2010Based Prediction of Protein\u2013Nucleic Acid Binding Using Graph Neural Networks","volume":"16","author":"Sagendorf J. M.","year":"2024","journal-title":"Biophysical Reviews"},{"key":"e_1_2_8_14_1","doi-asserted-by":"publisher","DOI":"10.1093\/nar\/gkab044"},{"issue":"1","key":"e_1_2_8_15_1","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbad488","article-title":"Protein\u2013DNA Binding Sites Prediction Based on Pre\u2010Trained Protein Language Model and Contrastive Learning","volume":"25","author":"Liu Y.","year":"2024","journal-title":"Briefings in Bioinformatics"},{"key":"e_1_2_8_16_1","doi-asserted-by":"crossref","DOI":"10.1016\/j.ijbiomac.2024.136147","article-title":"PDNAPred: Interpretable Prediction of Protein\u2013DNA Binding Sites Based on Pre\u2010Trained Protein Language Models","volume":"281","author":"Zhang L.","year":"2024","journal-title":"International Journal of Biological Macromolecules"},{"issue":"5","key":"e_1_2_8_17_1","doi-asserted-by":"crossref","DOI":"10.1093\/nar\/gkae039","article-title":"EquiPNAS: Improved Protein\u2013Nucleic Acid Binding Site Prediction Using Protein\u2010Language\u2010Model\u2010Informed Equivariant Deep Graph Neural Networks","volume":"52","author":"Roche R.","year":"2024","journal-title":"Nucleic Acids Research"},{"key":"e_1_2_8_18_1","doi-asserted-by":"crossref","DOI":"10.1016\/j.jmb.2023.168091","article-title":"LigBind: Identifying Binding Residues for Over 1000 Ligands With Relation\u2010Aware Graph Neural Networks","volume":"435","author":"Xia Y.","year":"2023","journal-title":"Journal of Molecular Biology"},{"issue":"1","key":"e_1_2_8_19_1","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbaf016","article-title":"Twenty Years of Advances in Prediction of Nucleic Acid\u2010Binding Residues in Protein Sequences","volume":"26","author":"Basu S.","year":"2025","journal-title":"Briefings in Bioinformatics"},{"key":"e_1_2_8_20_1","doi-asserted-by":"publisher","DOI":"10.1126\/science.ads0018"},{"key":"e_1_2_8_21_1","doi-asserted-by":"crossref","first-page":"1123","DOI":"10.1126\/science.ade2574","article-title":"Evolutionary\u2010Scale Prediction of Atomic\u2010Level Protein Structure With a Language Model","volume":"379","author":"Lin Z.","year":"2023","journal-title":"Science"},{"key":"e_1_2_8_22_1","doi-asserted-by":"crossref","unstructured":"S.Candido T.Hayes A.Derry et\u00a0al. Language Modeling Materializes a World Model of Protein Biology. BioRxiv: the Preprint Server for Biology(2026) https:\/\/doi.org\/10.64898\/2026.06.03.729735.","DOI":"10.64898\/2026.06.03.729735"},{"key":"e_1_2_8_23_1","first-page":"8844","volume-title":"Proceedings of International Conference on Machine Learning Research","author":"Rao R. M.","year":"2021"},{"key":"e_1_2_8_24_1","doi-asserted-by":"crossref","first-page":"173","DOI":"10.1038\/nmeth.1818","article-title":"HHblits: Lightning\u2010Fast Iterative Protein Sequence Searching by HMM\u2013HMM Alignment","volume":"9","author":"Remmert M.","year":"2012","journal-title":"Nature Methods"},{"key":"e_1_2_8_25_1","doi-asserted-by":"crossref","first-page":"3389","DOI":"10.1093\/nar\/25.17.3389","article-title":"Gapped BLAST and PSI\u2010BLAST: A New Generation of Protein Database Search Programs","volume":"25","author":"Altschul S. F.","year":"1997","journal-title":"Nucleic Acids Research"},{"key":"e_1_2_8_26_1","doi-asserted-by":"crossref","first-page":"177","DOI":"10.1007\/978-1-59745-514-5_10","article-title":"PSI\u2010BLAST Tutorial","volume":"395","author":"Bhagwat M.","year":"2007","journal-title":"Methods in Molecular Biology"},{"key":"e_1_2_8_27_1","doi-asserted-by":"crossref","first-page":"D20","DOI":"10.1093\/nar\/gkab1112","article-title":"Database Resources of the National Center for Biotechnology Information","volume":"50","author":"Sayers E. W.","year":"2022","journal-title":"Nucleic Acids Research"},{"key":"e_1_2_8_28_1","first-page":"47","article-title":"Swiss\u2010Prot: Connecting Biomolecular Knowledge via a Protein Database","volume":"3","author":"Gasteiger E.","year":"2001","journal-title":"Current Issues in Molecular Biology"},{"key":"e_1_2_8_29_1","doi-asserted-by":"crossref","first-page":"1282","DOI":"10.1093\/bioinformatics\/btm098","article-title":"UniRef: Comprehensive and Non\u2010Redundant UniProt Reference Clusters","volume":"23","author":"Suzek B. E.","year":"2007","journal-title":"Bioinformatics"},{"key":"e_1_2_8_30_1","doi-asserted-by":"crossref","first-page":"D170","DOI":"10.1093\/nar\/gkw1081","article-title":"Uniclust Databases of Clustered and Deeply Annotated Protein Sequences and Alignments","volume":"45","author":"Mirdita M.","year":"2017","journal-title":"Nucleic Acids Research"},{"key":"e_1_2_8_31_1","doi-asserted-by":"publisher","DOI":"10.7554\/eLife.93695"},{"key":"e_1_2_8_32_1","doi-asserted-by":"crossref","first-page":"2577","DOI":"10.1002\/bip.360221211","article-title":"Dictionary of Protein Secondary Structure: Pattern Recognition of Hydrogen\u2010Bonded and Geometrical Features","volume":"22","author":"Kabsch W.","year":"1983","journal-title":"Biopolymers"},{"key":"e_1_2_8_33_1","unstructured":"B.Jing S.Eismann P.Suriana R. J. L.Townshend andR. O.Dror Learning From Protein Structure With Geometric Vector Perceptrons. International Conference on Learning Representations(2021)."},{"key":"e_1_2_8_34_1","doi-asserted-by":"crossref","first-page":"583","DOI":"10.1038\/s41586-021-03819-2","article-title":"Highly Accurate Protein Structure Prediction With AlphaFold","volume":"596","author":"Jumper J.","year":"2021","journal-title":"Nature"},{"key":"e_1_2_8_35_1","doi-asserted-by":"crossref","DOI":"10.1093\/bib\/bbab564","article-title":"AlphaFold2\u2010Aware Protein\u2013DNA Binding Site Prediction Using Graph Transformer","volume":"23","author":"Yuan Q.","year":"2022","journal-title":"Briefings in Bioinformatics"},{"key":"e_1_2_8_36_1","doi-asserted-by":"publisher","DOI":"10.1093\/bib\/bbaf524"},{"key":"e_1_2_8_37_1","doi-asserted-by":"crossref","first-page":"7838","DOI":"10.1038\/s41467-024-52293-7","article-title":"Improving Prediction Performance of General Protein Language Model by Domain\u2010Adaptive Pretraining on DNA\u2010Binding Protein","volume":"15","author":"Zeng W.","year":"2024","journal-title":"Nature Communications"},{"key":"e_1_2_8_38_1","first-page":"76","volume-title":"IEEE International Conference on Bioinformatics and Biomedicine","author":"Zeng W.","year":"2023"},{"key":"e_1_2_8_39_1","doi-asserted-by":"crossref","first-page":"7112","DOI":"10.1109\/TPAMI.2021.3095381","article-title":"ProtTrans: Toward Understanding the Language of Life Through Self\u2010Supervised Learning","volume":"44","author":"Elnaggar A.","year":"2022","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"e_1_2_8_40_1","doi-asserted-by":"crossref","DOI":"10.1073\/pnas.2016239118","article-title":"Biological Structure and Function Emerge From Scaling Unsupervised Learning to 250 Million Protein Sequences","volume":"118","author":"Rives A.","year":"2021","journal-title":"Proceedings. National Academy of Sciences. United States of America"},{"key":"e_1_2_8_41_1","doi-asserted-by":"crossref","first-page":"493","DOI":"10.1038\/s41586-024-07487-w","article-title":"Accurate Structure Prediction of Biomolecular Interactions With AlphaFold 3","volume":"630","author":"Abramson J.","year":"2024","journal-title":"Nature"},{"key":"e_1_2_8_42_1","doi-asserted-by":"publisher","DOI":"10.1101\/2025.06.14.659707"}],"container-title":["Proteins: Structure, Function, and Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/prot.70190","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/prot.70190","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/prot.70190","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T11:46:28Z","timestamp":1791459988000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1002\/prot.70190"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,8]]},"references-count":41,"alternative-id":["10.1002\/prot.70190"],"URL":"https:\/\/doi.org\/10.1002\/prot.70190","archive":["Portico"],"relation":{},"ISSN":["0887-3585","1097-0134"],"issn-type":[{"value":"0887-3585","type":"print"},{"value":"1097-0134","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,8]]},"assertion":[{"value":"2025-12-13","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-09-28","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-10-08","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"prot.70190"}}