{"database":"biostudies-literature","file_versions":[],"scores":null,"additional":{"submitter":["Lu K"],"funding":["NIGMS NIH HHS"],"pagination":["138-151"],"full_dataset_link":["https://www.ebi.ac.uk/biostudies/studies/S-EPMC12744964"],"repository":["biostudies-literature"],"omics_type":["Unknown"],"volume":["66(1)"],"pubmed_abstract":["Natural products, synthesized via enzymes encoded by biosynthetic gene clusters (BGCs), represent a major source of therapeutic agents. Accurate BGC annotation is essential to unlocking the vast potential of natural product diversity. However, BGC annotation remains challenging due to our incomplete understanding of the enzymatic logic underlying biosynthesis. Here, we present two deep learning models trained on experimentally validated BGC-natural product pairs to advance BGC annotation. The BGC-multihead attention classifier (BGC-MAC) classifies BGCs by natural product class, outperforming antiSMASH and DeepBGC. The BGC-multihead attention product-matcher (BGC-MAP) associates BGCs with product structures, demonstrating potential to prioritize candidate BGCs given a natural product or to "],"journal":["Journal of chemical information and modeling"],"pubmed_title":["BGC-MAC and BGC-MAP: Attention-Based Models for Biosynthetic Gene Cluster Classification and Product Matching."],"pmcid":["PMC12744964"],"funding_grant_id":["R35 GM128742","RM1 GM145426","R35 GM159995"],"pubmed_authors":["Lu K","Ding Y","Geng H","Chen M","Fu TM","Luesch H","Xu W","Li M","Xie WJ"],"additional_accession":[]},"is_claimable":false,"name":"BGC-MAC and BGC-MAP: Attention-Based Models for Biosynthetic Gene Cluster Classification and Product Matching.","description":"Natural products, synthesized via enzymes encoded by biosynthetic gene clusters (BGCs), represent a major source of therapeutic agents. Accurate BGC annotation is essential to unlocking the vast potential of natural product diversity. However, BGC annotation remains challenging due to our incomplete understanding of the enzymatic logic underlying biosynthesis. Here, we present two deep learning models trained on experimentally validated BGC-natural product pairs to advance BGC annotation. The BGC-multihead attention classifier (BGC-MAC) classifies BGCs by natural product class, outperforming antiSMASH and DeepBGC. The BGC-multihead attention product-matcher (BGC-MAP) associates BGCs with product structures, demonstrating potential to prioritize candidate BGCs given a natural product or to ","dates":{"release":"2026-01-01T00:00:00Z","publication":"2026 Jan","modification":"2026-06-06T19:50:28.908Z","creation":"2026-06-04T03:12:40.274Z"},"accession":"S-EPMC12744964","cross_references":{"pubmed":["41412814"],"doi":["10.1021/acs.jcim.5c02082"]}}