Joint enzyme-reaction retrieval and catalytic optima prediction via multimodal fusion
Preprint
Abstract
Motivation: Enzyme-reaction retrieval is increasingly used to prioritize candidate biocatalysts for experimental follow-up, where useful recommendations should indicate not only whether an enzyme can catalyze a target reaction but also under which pH and temperature conditions it should be tested. Existing retrieval models optimize catalytic matching scores, whereas catalytic optima predictors are typically developed as enzyme-level regressors because public pH and temperature annotations are sparse and often available only at the enzyme or EC-associated record level. This separation leaves a practical gap: high-ranking enzyme-reaction pairs are not evaluated for condition suitability, and enzyme-level optima predictions do not use the reaction context being retrieved. Results: We present GERO, a multimodal fusion framework that uses feature-gated cross-modal fusion to integrate global enzyme sequence semantics, sequence-derived pocket geometry, and molecular reaction representations for condition-aware enzyme-reaction retrieval and catalytic optima estimation with reaction context. To evaluate this setting, we define the tolerance-restricted hit rate (Hit@k-TR), which requires both top-k retrieval of the correct candidate and condition prediction within predefined tolerances. Across enzyme- and reaction-similarity splits, GERO improves Hit@k-TR over two-stage retrieval-then-prediction baselines. Representative benchmark examples and an iodinin biosynthesis case study further illustrate GERO’s ability to provide candidate rankings together with plausible assay-condition estimates for downstream experimental prioritization. Availability and implementation: Source code is available at https://github.com/ykxhs/GERO . Contact: liujuan@whu.edu.cn Supplementary information: Supplementary data are available at XXXX online.Keywords
Funding Information
- National Key Research and Development Program of China (2019YFA0904303)
- Intelligent Computing Center of the National Cybersecurity Talent and Innovation Base, Wuhan
This publication has 17 references indexed in Scilit:
- Enzyme specificity prediction using cross-attention graph neural networksNature, 2025
- Machine learning prediction of enzyme optimum pHNature Machine Intelligence, 2025
- Seq2Topt: a sequence-based deep learning predictor of enzyme optimal temperatureBriefings in Bioinformatics, 2025
- Highly accurate protein structure prediction with AlphaFoldNature, 2021
- Mapping the space of chemical reactions using attention-based neural networksNature Machine Intelligence, 2021
- Dual phenazine gene clusters enable diversification during biosynthesisNature Chemical Biology, 2019
- TaggerOne: joint named entity recognition and normalization with semi-Markov ModelsBioinformatics, 2016
- Heterocyclic Aromatic N-Oxidation in the Biosynthesis of Phenazine Antibiotics from Lysobacter antibioticusOrganic Letters, 2016
- Industrial enzyme applicationsCurrent Opinion in Biotechnology, 2002
- The Effect of Temperature and pH on Enzyme Activity1975