Uncovering Microbial Biosynthetic Potential with Genomic Context-aware Protein Language Model

2025/05/03 by Zixin Kang, Haohong Zhang, Chaoqin Liang +5 · 1 voice
Biochemistry, Genetics and Molecular Biology · #Biomedical Text Mining and Ontologies #Genomics and Phylogenetic Studies #Machine Learning in Bioinformatics

paper · pdf · doi:10.1101/2025.04.29.651206

openalex publication_date 2025/05/03 · openalex created_date 2025/05/04 · openalex updated_date 2026/07/14

Abstract

Abstract Microbial secondary metabolites, synthesized by biosynthetic gene clusters (BGCs), offer vast potential for biotechnological applications. Among BGC profiling techniques, computational detection methods face challenges, including time-consuming alignment and reliance on predefined profiles. To address these, we present BGC-Finder, an end-to-end pipeline utilizing protein language models for BGC detection and annotation from microbial genomes and metagenomes. This approach achieves remarkable increase in profiling speed of up to 100-fold, and employs genomic context-aware modeling to facilitate interpretable genetic essentiality assessment and large-scale BGC clustering. BGC-Finder outperformed traditional methods, successfully detecting 9.49% more biosynthetic-core genes and 27.70% more cytochrome P450s in 742 experimentally-validated BGCs. Notably, it retrieved 31 remote biosynthetic homologs from 210 polar marine metagenomes and identified 4,585 BGCs with 6,388 core genes from 256 fungal genomes. These findings highlight BGC-Finder’s capability to illuminate “microbial biosynthesis dark matter” (sequence-unrelated, function-similar biosynthetic enzymes) and expedite natural product discovery. Highlights BGC-Finder is an accurate and ultrafast pipeline leveraging protein language models (pLMs) to predict and annotate biosynthetic gene clusters (BGCs) from microbial genomes and metagenomes. The genomic context-aware model enables interpretable analysis: attention-driven identification of essential biosynthetic genes and embedding-guided BGC clustering. BGC-Finder sensitively retrieves remote homologous BGCs from both bacteria and fungi genomes, uncovering hidden ‘microbial biosynthesis dark matter’. We discovered a non-ribosomal peptide synthetase (NRPS) family, which involved into function-specific BGCs in two evolutionarily distant fungi.

Citations

• Improve Transformer Models with Better Relative Position Embeddings
• Highly accurate protein structure prediction with AlphaFold
• Large language models generate functional protein sequences across diverse families
• Evolutionary-scale prediction of atomic-level protein structure with a language model
• An Arabic Auto-indexing System for Information Retrieval.
• Interactive Tree of Life (iTOL) v6: recent updates to the phylogenetic tree display and annotation tool
• Accurate structure prediction of biomolecular interactions with AlphaFold 3
• Enzymatic combinatorial synthesis of E-64 and related cysteine protease inhibitors
• Progress and Opportunities of Foundation Models in Bioinformatics
• Fast and accurate protein structure search with Foldseek
• UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction

Discussions

• Deciphering Biosynthetic Gene Clusters with a Context-aware Protein Language Model doi.org/10.1101/2025... Phylogenomics and metabolic engineering reveal a conserved gene cluster in Solanaceae plants [bsky, 0 points, 0 comments]

Related

• A novel transformer-based platform for the prediction and design of biosynthetic gene clusters for (un)natural products
• The secondary metabolism collaboratory: a database and web discussion portal for secondary metabolite biosynthetic gene clusters
• High-throughput prediction of protein–protein interactions uncovers hidden molecular networks in biosynthetic gene clusters
• Phylogeny‐Aware Metabologenomics Accurately Assigns Natural Products to Biosynthetic Gene Clusters
• Enabling Access to Novel Bacterial Biosynthetic Potential From ONT Draft Genomic Data