{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Zhu S"],"funding":["National Institute of Allergy and Infectious Diseases","National Institute of General Medical Sciences","NIGMS NIH HHS","National Science Foundation"],"pagination":["e70340"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC12617271"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["34(12)"],"pubmed_abstract":["Computational design of functional proteins is of both fundamental and applied interest. This study introduces a generative framework for co-designing protein sequence and structure in a unified process by modeling their joint distribution, with the goal of enabling cross-modality interactions toward coherent and functional designs. Each residue is represented by three distinct modalities (type, position, and orientation) and modeled using dedicated diffusion processes: multinomial for types, Cartesian for positions, and special orthogonal group SO(3) for orientations. To couple these modalities, we propose a unified architecture, ReverseNet, which employs a shared graph attention encoder to integrate multimodal information and separate projectors to predict each modality. We benchmark our"],"journal":["Protein science : a publication of the Protein Society"],"pubmed_title":["Multimodal diffusion for joint design of protein sequence and structure."],"pmcid":["PMC12617271"],"funding_grant_id":["R35GM124952","1943008","R01AI165433","2203715","2514387"],"pubmed_authors":["Liu Y","Sun Q","Kotnis S","Shen Y","Gulati S","Zhu S"],"additional_accession":[]},"is_claimable":false,"name":"Multimodal diffusion for joint design of protein sequence and structure.","description":"Computational design of functional proteins is of both fundamental and applied interest. This study introduces a generative framework for co-designing protein sequence and structure in a unified process by modeling their joint distribution, with the goal of enabling cross-modality interactions toward coherent and functional designs. Each residue is represented by three distinct modalities (type, position, and orientation) and modeled using dedicated diffusion processes: multinomial for types, Cartesian for positions, and special orthogonal group SO(3) for orientations. To couple these modalities, we propose a unified architecture, ReverseNet, which employs a shared graph attention encoder to integrate multimodal information and separate projectors to predict each modality. We benchmark our","dates":{"release":"2025-01-01T00:00:00Z","publication":"2025 Dec","modification":"2026-06-05T14:36:38.84Z","creation":"2026-05-18T03:07:58.789Z"},"accession":"S-EPMC12617271","cross_references":{"pubmed":["41236386"],"doi":["10.1002/pro.70340"]}}