文献类型：期刊文献

英文题名：A multimodal framework for comprehensive driver variant prediction in cancer

作者：Yang, Hai[1,2];Chen, Yijia[2];Zhou, Tianyi[3];Wang, Yingzhuo[4];Zhou, Qin[1,2];Xiao, Ting[1,2];Zhang, Qian[1,2];Zhang, Jing[1,2];Li, Dongdong[1,2];Wang, Zhe[1,2]

机构：[1]East China Univ Sci & Technol, Key Lab Smart Mfg Energy Chem Proc, Minist Educ, Shanghai, Peoples R China;[2]East China Univ Sci & Technol, Dept Comp Sci & Engn, Shanghai, Peoples R China;[3]Univ Michigan, Elect Engn & Comp Sci Dept, Ann Arbor, MI USA;[4]Imperial Coll London, Fac Engn, London, England

年份：2025

卷号：5

期号：1

外文期刊名：COMMUNICATIONS MEDICINE

收录：WOS:【ESCI(收录号:WOS:001624930900001)】；

基金：This work was supported by the National Key Research and Development Program of China under Grant 2023YFF1204904, Natural Science Foundation of China under Grant No. 61902126, Shanghai Science and Technology Program "Federated based cross-domain and cross-task incremental learning" under Grant No. 21511100800.

语种：英文

摘要：BackgroundCancer genomes contain many mutations, but only a subset drive tumor development. Accurately pinpointing these driver variants remains challenging. We aim to build an accurate and interpretable model by combining DNA sequence, protein 3D structure, and cancer omics data.MethodsWe present ModVAR, a multimodal model that integrates DNA sequences, predicted protein tertiary structures, and cancer omics data to classify driver variants. The approach uses pre-trained models (DNAbert2 and ESMFold) and a self-supervised strategy for cancer omics profiles. We evaluate performance on clinically and experimentally validated driver variants with standard classification metrics, examine therapeutic relevance through molecular docking, assess modeling of variants in intrinsically disordered protein regions, and analyze modality contributions.ResultsHere we show that ModVAR achieves strong accuracy across benchmarks for identifying validated driver variants. It prioritizes variants with potential therapeutic actionability supported by docking analyses, and the inclusion of structural predictions enables effective modeling of variants in intrinsically disordered regions. Interpretation indicates that the protein structure modality contributes most to predictions. At scale, the method produces 3,971,946 publicly available variant annotations.ConclusionsModVAR integrates sequence, structure, and cancer omics signals to aid driver-variant discovery. It provides robust performance across tasks, supports hypothesis generation and target discovery, and supplies a large-scale resource that advances cancer research and personalized therapy.