Prediction of bacteria-specific drug metabolism using a cross-attention-based model.
6 affiliations
- 1Drug Information Platform Center, Chemical Platform Technology Division, Korea Research Institute of Chemical Technology (KRICT), 141, Gajeong-ro, Yuseong-gu, Daejeon 34114, Republic of Korea.
- 2Department of Medicinal Chemistry and Pharmacology, University of Science and Technology (UST), 217, Gajeong-ro, Yuseong-gu, Daejeon 34113, Republic of Korea.
- 3Chemical Analysis Center, Chemical Platform Technology Division, Korea Research Institute of Chemical Technology (KRICT), 141, Gajeong-ro, Yuseong-gu, Daejeon 34114, Republic of Korea.
- 4Center for Biomimetic Research, Korea Institute of Toxicology (KIT), 141, Gajeong-ro, Yuseong-gu, Daejeon 34114, Republic of Korea.
- 5School of Artificial Intelligence, Chonnam National University, 77 Yongbong-ro, Buk-gu, Jeonnam Gwangju 61186, Republic of Korea.
- 6Department of AI-Based Drug Discovery, Duksung Women's University, 33, Samyang-ro 144-gil, Dobonggu Seoul 01369, Republic of Korea.
Abstract
Drug metabolism by gut microorganisms is bacteria-specific and can have clinically significant consequences, including prodrug activation, drug inactivation, and the generation of toxic metabolites. However, existing computational approaches have primarily focused on community-level predictions or drug-to-microbe effects, without resolving the metabolic contributions of individual bacterial strains or the bidirectional nature of drug-bacteria interactions, despite advances in computational frameworks. Here, we propose a bidirectional cross-attention architecture, multimodal Bi-MHA, that integrates atom-level molecular embeddings and protein-level bacterial embeddings from pretrained models to generate a combined representation for classification. The proposed model was developed using reliability stratification and a λ-scaled loss function on the training dataset and achieved consistent performance across three unseen tasks: unseen drug (receiver operating characteristic area under the curve [ROC AUC]: 0.690), unseen bacteria (ROC AUC: 0.920), and unseen interaction (ROC AUC: 0.949), ranking first among all compared models in the rank-sum metric. Beyond prediction, attention-based analysis not only indicated enzymatic reaction sites that were experimentally validated in previous studies but also prioritized candidate bacterial proteins, demonstrating interpretability at both the atom and protein levels. The proposed model also achieved a 15 percentage point improvement in accuracy over the KNN baseline on an external validation dataset of 85 out-of-distribution drug-bacteria pairs. Furthermore, the model was extended to community-level prediction as a proof-of-concept application using a weighted-abundance pipeline on an independent microbiome dataset. These results demonstrate the model's potential as an interpretable screening tool for drug safety evaluation and personalized medicine. The source code is available at https://github.com/Jaehyeon-O-Ob/Bacteria-Specific_Drug_Metabolism.
Related
- Cross-tissue graph attention networks for semi-supervised gene expression prediction
- Pep2Net: an integrated peptide knowledge base and cross-attention model for multi-task prediction of peptides
- Functional properties of specific peptides derived from whey protein fermented with lactic acid bacteria isolated from dog feces.
- Colon-Specific Gut Microbiome Remodelling Is Associated with Host Metabolism in Paracetamol-Induced Liver Injury in Rats
- Myofiber-specific knockout of TGF-β type I receptors in mice concurrently drives muscle hypertrophy, oxidative metabolism, and absolute force
BibTeX
@article{park2026prediction,
title = {{Prediction of bacteria-specific drug metabolism using a cross-attention-based model.}},
author = {Jaehyeon Park and Si Eun Kim and Heeyoung Yang and Nam-Chul Cho and Hyunho Kim and Hyun Kil Shin},
journal = {Briefings in bioinformatics},
year = {2026},
month = sep,
doi = {10.1093/bib/bbag578},
url = {https://europepmc.org/article/MED/42894386},
note = {CC BY},
}
RIS
TY - JOUR
TI - Prediction of bacteria-specific drug metabolism using a cross-attention-based model.
AU - Jaehyeon Park
AU - Si Eun Kim
AU - Heeyoung Yang
AU - Nam-Chul Cho
AU - Hyunho Kim
AU - Hyun Kil Shin
PY - 2026
DA - 2026/09/01
JO - Briefings in bioinformatics
DO - 10.1093/bib/bbag578
UR - https://europepmc.org/article/MED/42894386
AB - Drug metabolism by gut microorganisms is bacteria-specific and can have clinically significant consequences, including prodrug activation, drug inactivation, and the generation of toxic metabolites. However, existing computational approaches have primarily focused on community-level predictions or drug-to-microbe effects, without resolving the metabolic contributions of individual bacterial strains or the bidirectional nature of drug-bacteria interactions, despite advances in computational frameworks. Here, we propose a bidirectional cross-attention architecture, multimodal Bi-MHA, that integrates atom-level molecular embeddings and protein-level bacterial embeddings from pretrained models to generate a combined representation for classification. The proposed model was developed using reliability stratification and a λ-scaled loss function on the training dataset and achieved consistent performance across three unseen tasks: unseen drug (receiver operating characteristic area under the curve [ROC AUC]: 0.690), unseen bacteria (ROC AUC: 0.920), and unseen interaction (ROC AUC: 0.949), ranking first among all compared models in the rank-sum metric. Beyond prediction, attention-based analysis not only indicated enzymatic reaction sites that were experimentally validated in previous studies but also prioritized candidate bacterial proteins, demonstrating interpretability at both the atom and protein levels. The proposed model also achieved a 15 percentage point improvement in accuracy over the KNN baseline on an external validation dataset of 85 out-of-distribution drug-bacteria pairs. Furthermore, the model was extended to community-level prediction as a proof-of-concept application using a weighted-abundance pipeline on an independent microbiome dataset. These results demonstrate the model's potential as an interpretable screening tool for drug safety evaluation and personalized medicine. The source code is available at https://github.com/Jaehyeon-O-Ob/Bacteria-Specific_Drug_Metabolism.
SN - 1467-5463
SN - 1477-4054
ER -