{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,26]],"date-time":"2026-03-26T16:17:50Z","timestamp":1774541870831,"version":"3.50.1"},"reference-count":44,"publisher":"Oxford University Press (OUP)","issue":"5","license":[{"start":{"date-parts":[[2025,10,26]],"date-time":"2025-10-26T00:00:00Z","timestamp":1761436800000},"content-version":"vor","delay-in-days":56,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,8,31]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Ribonucleic acids (RNAs) play a central role in cellular processes by interacting with proteins, small molecules, and other RNAs. Accurate prediction of these interactions is critical for understanding post-transcriptional regulation and advancing RNA-targeted therapeutics. However, existing computational methods are limited by their reliance on hand-crafted features, modality-specific architectures, and often require structural or physicochemical data, which are experimentally challenging to obtain and unavailable for many RNA molecules. These constraints hinder generalizability and fail to capture the complex, context-dependent semantics of RNA interactions. We present BioLLMNet, a unified sequence-only framework that leverages pretrained biological language models to encode rich, contextualized representations for both RNA molecules and their interacting partners, including proteins, small molecules, and other RNAs. Our key innovation is the introduction of a novel learnable gating mechanism, which dynamically computes feature-wise weights to adaptively integrate multimodal embeddings based on input context. This mechanism, proposed here for the first time in RNA interaction modeling, enables the model to emphasize the most informative features from each partner and achieves seamless fusion of heterogeneous modalities. As a result, BioLLMNet represents a unified framework and can flexibly and consistently model all three types of interaction (RNA\u2013protein, RNA\u2013small molecule, and RNA\u2013RNA) within a shared architecture, eliminating the need for modality-specific designs. Comprehensive evaluations on benchmark data sets demonstrate that BioLLMNet achieves state-of-the-art performance across all three types of interaction. Our results underscore the power of language model-based representations combined with dynamic feature fusion for generalizable, modality-aware RNA interaction prediction.<\/jats:p>","DOI":"10.1093\/bib\/bbaf549","type":"journal-article","created":{"date-parts":[[2025,10,2]],"date-time":"2025-10-02T11:59:25Z","timestamp":1759406365000},"source":"Crossref","is-referenced-by-count":1,"title":["BioLLMNet: enhancing RNA-interaction prediction with a specialized cross-LLM transformation network"],"prefix":"10.1093","volume":"26","author":[{"given":"Abrar Rahman","family":"Abir","sequence":"first","affiliation":[{"name":"Department of Computer Science and Engineering , Bangladesh University of Engineering and Technology, Dhaka 1000,","place":["Bangladesh"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Md","family":"Toki Tahmid","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering , Bangladesh University of Engineering and Technology, Dhaka 1000,","place":["Bangladesh"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Md Shamsuzzoha","family":"Bayzid","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering , Bangladesh University of Engineering and Technology, Dhaka 1000,","place":["Bangladesh"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2025,10,26]]},"reference":[{"key":"2025102609214063400_ref1","doi-asserted-by":"publisher","first-page":"33","DOI":"10.1186\/s12943-022-01514-y","article-title":"CircPVT1: a pivotal circular node intersecting long non-coding-PVT1 and c-MYC oncogenic signals","volume":"21","author":"Palcau","year":"2022","journal-title":"Mol Cancer"},{"key":"2025102609214063400_ref2","doi-asserted-by":"publisher","first-page":"397","DOI":"10.1093\/nar\/gkab1208","article-title":"Identification and targeting of g-quadruplex structures in MALAT1 long non-coding","volume":"50","author":"Mou","year":"2022","journal-title":"Nucleic Acids Res"},{"key":"2025102609214063400_ref3","doi-asserted-by":"publisher","first-page":"475","DOI":"10.1038\/s41580-020-0243-y","article-title":"The expanding regulatory mechanisms and cellular functions of circular RNAs","volume":"21","author":"Chen","year":"2020","journal-title":"Nat Rev Mol Cell Biol"},{"key":"2025102609214063400_ref4","doi-asserted-by":"publisher","first-page":"22","DOI":"10.1038\/s41568-020-00306-0","article-title":"RNA in cancer","volume":"21","author":"Goodall","year":"2021","journal-title":"Nat Rev Cancer"},{"key":"2025102609214063400_ref5","doi-asserted-by":"publisher","first-page":"225","DOI":"10.1038\/s41592-019-0330-1","article-title":"Methods to study RNA\u2013protein interactions","volume":"16","author":"Ramanathan","year":"2019","journal-title":"Nat Methods"},{"key":"2025102609214063400_ref6","doi-asserted-by":"publisher","first-page":"e52","DOI":"10.1093\/nar\/gkaa143","article-title":"Capturing RNA\u2013protein interaction via CRUIS","volume":"48","author":"Zhang","year":"2020","journal-title":"Nucleic Acids Res"},{"key":"2025102609214063400_ref7","doi-asserted-by":"publisher","first-page":"184","DOI":"10.1038\/s41592-019-0666-6","article-title":"Deciphering interaction fingerprints from protein molecular surfaces using geometric deep learning","volume":"17","author":"Gainza","year":"2020","journal-title":"Nat Methods"},{"key":"2025102609214063400_ref8","doi-asserted-by":"publisher","first-page":"1137","DOI":"10.1093\/bib\/bbab062","article-title":"The miRNA: a small but powerful RNA for covid-19","volume":"22","author":"Zhang","year":"2021","journal-title":"Brief Bioinform"},{"key":"2025102609214063400_ref9","doi-asserted-by":"publisher","first-page":"791","DOI":"10.1016\/j.cell.2018.07.023","article-title":"Emerging roles for intermolecular RNA\u2013RNA interactions in RNP assemblies","volume":"174","author":"Van Treeck","year":"2018","journal-title":"Cell"},{"key":"2025102609214063400_ref10","doi-asserted-by":"publisher","first-page":"2986","DOI":"10.1093\/bioinformatics\/btaa074","article-title":"PmliPred: a method based on hybrid model and fuzzy decision for plant miRNA-lncRNA interaction prediction","volume":"36","author":"Kang","year":"2020","journal-title":"Bioinformatics"},{"key":"2025102609214063400_ref11","doi-asserted-by":"publisher","first-page":"347","DOI":"10.1038\/s41587-020-0709-7","article-title":"Analysis of RNA\u2013protein networks with RNP-map defines functional hubs on RNA","volume":"39","author":"Weidmann","year":"2021","journal-title":"Nat Biotechnol"},{"key":"2025102609214063400_ref12","doi-asserted-by":"publisher","first-page":"444","DOI":"10.1038\/nmeth.1611","article-title":"Predicting protein associations with long noncoding RNAs","volume":"8","author":"Bellucci","year":"2011","journal-title":"Nat Methods"},{"key":"2025102609214063400_ref13","doi-asserted-by":"publisher","first-page":"104611","DOI":"10.1016\/j.compbiomed.2021.104611","article-title":"High-throughput analysis of the interactions between viral proteins and host cell RNAs","volume":"135","author":"Lanjanian","year":"2021","journal-title":"Comput Biol Med"},{"key":"2025102609214063400_ref14","doi-asserted-by":"publisher","first-page":"e107","DOI":"10.1093\/nar\/gkab614","article-title":"Visualization and characterization of RNA\u2013protein interactions in living cells","volume":"49","author":"Duan","year":"2021","journal-title":"Nucleic Acids Res"},{"key":"2025102609214063400_ref15","doi-asserted-by":"publisher","first-page":"502","DOI":"10.1038\/nrc.2017.36","article-title":"Drugging the \u2018undruggable\u2019 cancer targets","volume":"17","author":"Dang","year":"2017","journal-title":"Nat Rev Cancer"},{"key":"2025102609214063400_ref16","doi-asserted-by":"publisher","first-page":"103756","DOI":"10.1016\/j.ebiom.2021.103756","article-title":"Therapeutic targeting of \u201cundruggable\u201d myc","volume":"75","author":"Llombart","year":"2022","journal-title":"EBioMedicine"},{"key":"2025102609214063400_ref17","doi-asserted-by":"publisher","first-page":"bbab046","DOI":"10.1093\/bib\/bbab046","article-title":"Predtis: prediction of drug-target interactions based on multiple feature information using gradient boosting framework with data balancing and feature selection techniques","volume":"22","author":"Mahmud","year":"2021","journal-title":"Brief Bioinform"},{"key":"2025102609214063400_ref18","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s12276-018-0071-8","article-title":"Single-cell RNA sequencing technologies and bioinformatics pipelines","volume":"50","author":"Hwang","year":"2018","journal-title":"Exp Mol Med"},{"key":"2025102609214063400_ref19","doi-asserted-by":"crossref","first-page":"1724","DOI":"10.1109\/TCBB.2020.3034922","article-title":"MD-MLI: prediction of miRNA-lncRNA interaction by using multiple features and hierarchical deep learning","volume":"19","author":"Song","year":"2020","journal-title":"IEEE\/ACM Trans Comput Biol Bioinform"},{"key":"2025102609214063400_ref20","doi-asserted-by":"crossref","first-page":"bbaa228","DOI":"10.1093\/bib\/bbaa228","article-title":"Predicting the interaction biomolecule types for lncrna: an ensemble deep learning approach","volume":"22","author":"Zhang","year":"2021","journal-title":"Brief Bioinform"},{"key":"2025102609214063400_ref21","doi-asserted-by":"publisher","first-page":"e110","DOI":"10.1093\/nar\/gkad929","article-title":"A task-specific encoding algorithm for RNAs and RNA-associated interactions based on convolutional autoencoder","volume":"51","author":"Wang","year":"2023","journal-title":"Nucleic Acids Res"},{"key":"2025102609214063400_ref22","doi-asserted-by":"publisher","first-page":"415","DOI":"10.1186\/s12859-021-04328-9","article-title":"PRPI-SC: an ensemble deep learning model for predicting plant lncRNA\u2013protein interactions","volume":"22","author":"Zhou","year":"2021","journal-title":"BMC Bioinform"},{"key":"2025102609214063400_ref23","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/1471-2105-12-489","article-title":"Predicting RNA-protein interactions using only sequence information","volume":"12","author":"Muppirala","year":"2011","journal-title":"BMC Bioinform"},{"key":"2025102609214063400_ref24","doi-asserted-by":"publisher","first-page":"e1544","DOI":"10.1002\/wrna.1544","article-title":"Recent methodology progress of deep learning for RNA\u2013protein interaction prediction","volume":"10","author":"Pan","year":"2019","journal-title":"Wiley Interdisc Rev RNA"},{"key":"2025102609214063400_ref25","doi-asserted-by":"publisher","first-page":"18","DOI":"10.3389\/fgene.2019.00018","article-title":"Construction of complex features for computational predicting ncRNA-protein interaction","volume":"10","author":"Dai","year":"2019","journal-title":"Front Genet"},{"key":"2025102609214063400_ref26","doi-asserted-by":"publisher","first-page":"1070","DOI":"10.3390\/ijms20051070","article-title":"RPITER: a hierarchical deep learning framework for ncRNA\u2013protein interaction prediction","volume":"20","author":"Peng","year":"2019","journal-title":"Int J Mol Sci"},{"key":"2025102609214063400_ref27","doi-asserted-by":"publisher","first-page":"107718","DOI":"10.1016\/j.compbiolchem.2022.107718","article-title":"LPI-CSFFR: combining serial fusion with feature reuse for predicting lncRNA-protein interactions","volume":"99","author":"Huang","year":"2022","journal-title":"Comput Biol Chem"},{"key":"2025102609214063400_ref28","doi-asserted-by":"publisher","first-page":"W509","DOI":"10.1093\/nar\/gkad404","article-title":"RNAincoder: a deep learning-based encoder for RNA and RNA-associated interaction","volume":"51","author":"Wang","year":"2023","journal-title":"Nucleic Acids Res"},{"key":"2025102609214063400_ref29","doi-asserted-by":"publisher","first-page":"1605","DOI":"10.1261\/rna.039834.113","article-title":"LigandRNA: computational predictor of RNA-ligand interactions","volume":"19","author":"Philips","year":"2013","journal-title":"RNA"},{"key":"2025102609214063400_ref30","doi-asserted-by":"publisher","first-page":"e78","DOI":"10.1093\/nar\/gkab356","article-title":"dSPRINT: predicting DNA, RNA, ion, peptide and small molecule interaction sites within protein domains","volume":"49","author":"Etzion-Fuchs","year":"2021","journal-title":"Nucleic Acids Res"},{"key":"2025102609214063400_ref31","doi-asserted-by":"crossref","first-page":"bbae002","DOI":"10.1093\/bib\/bbae002","article-title":"Reliable method for predicting the binding affinity of RNA-small molecule interactions using machine learning","volume":"25","author":"Krishnan","year":"2024","journal-title":"Brief Bioinform"},{"key":"2025102609214063400_ref32","doi-asserted-by":"publisher","first-page":"247","DOI":"10.1038\/s42003-025-07694-9","article-title":"RNA-protein interaction prediction using network-guided deep learning","volume":"8","author":"Liu","year":"2025","journal-title":"Communications Biology"},{"key":"2025102609214063400_ref33","doi-asserted-by":"crossref","DOI":"10.1038\/s41467-025-60872-5","article-title":"Rinalmo: General-purpose rna language models can generalize well on structure prediction tasks","volume-title":"Nature Communications","author":"Peni\u0107"},{"key":"2025102609214063400_ref34","article-title":"Language models of protein sequences at the scale of evolution enable accurate structure prediction","author":"Lin","year":"2022","journal-title":"BioRxiv"},{"key":"2025102609214063400_ref35","doi-asserted-by":"crossref","article-title":"Mole-bert: rethinking pre-training graph neural networks for molecules","author":"Xia","DOI":"10.26434\/chemrxiv-2023-dngg4"},{"key":"2025102609214063400_ref36","article-title":"LOCAS: multi-label mRNA localization with supervised contrastive learning","volume":"26","author":"Abir","journal-title":"Briefings in Bioinformatics"},{"key":"2025102609214063400_ref37","doi-asserted-by":"crossref","first-page":"bbaf199","DOI":"10.1093\/bib\/bbaf199","article-title":"DeepRNA-twist: language-model-guided RNA torsion angle prediction with attention-inception network","volume":"26","author":"Abir","year":"2025","journal-title":"Brief Bioinform"},{"key":"2025102609214063400_ref38","doi-asserted-by":"crossref","first-page":"btae533","DOI":"10.1093\/bioinformatics\/btae533","article-title":"DTI-LM: language model powered drug\u2013target interaction prediction","volume":"40","author":"Ahmed","year":"2024","journal-title":"Bioinformatics"},{"key":"2025102609214063400_ref39","doi-asserted-by":"publisher","first-page":"4310","DOI":"10.1021\/acs.jcim.3c02061","article-title":"Predicting antimicrobial peptides using ESMfold-predicted structures and ESM-2-based amino acid features with graph deep learning","volume":"64","author":"Cordoves-Delgado","year":"2024","journal-title":"J Chem Inf Model"},{"key":"2025102609214063400_ref40","article-title":"Transformers for molecular property prediction: domain adaptation efficiently improves performance","author":"Sultan","year":"2025"},{"key":"2025102609214063400_ref41","doi-asserted-by":"crossref","first-page":"btae155","DOI":"10.1093\/bioinformatics\/btae155","article-title":"Contrastive pre-training and 3d convolution neural network for RNA and small molecule binding affinity prediction","volume":"40","author":"Sun","year":"2024","journal-title":"Bioinformatics"},{"key":"2025102609214063400_ref42","doi-asserted-by":"publisher","first-page":"W72","DOI":"10.1093\/nar\/gkab393","article-title":"cat rapid omics v2. 0: going deeper and wider in the prediction of protein\u2013RNA interactions","volume":"49","author":"Armaos","year":"2021","journal-title":"Nucleic Acids Res"},{"key":"2025102609214063400_ref43","doi-asserted-by":"publisher","first-page":"e363","DOI":"10.1371\/journal.pbio.0020363","article-title":"Human microRNA targets","volume":"2","author":"John","year":"2004","journal-title":"PLoS Biol"},{"key":"2025102609214063400_ref44","first-page":"34892","article-title":"Visual instruction tuning","volume":"36","author":"Liu","year":"2023","journal-title":"Adv Neural Inform Process Syst"}],"container-title":["Briefings in Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/26\/5\/bbaf549\/64948694\/bbaf549.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bib\/article-pdf\/26\/5\/bbaf549\/64948694\/bbaf549.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,26]],"date-time":"2025-10-26T13:21:50Z","timestamp":1761484910000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bib\/article\/doi\/10.1093\/bib\/bbaf549\/8303308"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8,31]]},"references-count":44,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2025,8,31]]}},"URL":"https:\/\/doi.org\/10.1093\/bib\/bbaf549","relation":{},"ISSN":["1467-5463","1477-4054"],"issn-type":[{"value":"1467-5463","type":"print"},{"value":"1477-4054","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2025,9]]},"published":{"date-parts":[[2025,8,31]]},"article-number":"bbaf549"}}