Enzyme Function Prediction


Overview

  1. Enzyme functional classification using artificial intelligence. Trends in Biotechnology, 2025. [PDF].

  2. Comparative Assessment of Protein Large Language Models for Enzyme Commission Number Prediction. BMC Bioinformatics, 2025. [PDF].

Large Language Model-Based Methods

  1. Interpretable Kolmogorov-Arnold Networks for Enzyme Commission Number Prediction. bioRxiv, 2025. Model: ESM-1b + KAN, [PDF], [Code].

  2. ProtDETR: Interpretable Enzyme Function Prediction via Residue-Level Detection. bioRxiv, 2025. Model: ESM-1b + Self-Attention + CNN, [PDF], [Code].

  3. ProteinF3S: Boosting enzyme function prediction by fusing protein sequence, structure, and surface. Briefings in Bioinformatics, 2025. Model: ESM2 + CNN + transformer, [PDF], [Code].

  4. PhiGnet: Accurate prediction of protein function using statistics-informed graph networks. Nature Communications, 2024. Model: ESM-1b + GCN. [PDF] [Code]

  5. GraphEC: Accurately predicting enzyme functions through geometric graph learning on ESMFold-predicted structures. Nature Communications, 2024. Model: ProtTrans + GNN + Attention, [PDF], [Code1], [Code2].

  6. CPEC: Leveraging conformal prediction to annotate enzyme function space with limited false positives. PLoS Computational Biology, 2024. Model: ESM + GNN + Triplet Loss, [PDF], [Code].

  7. CLEAN-Contact: Improved enzyme functional annotation prediction using contrastive learning with structural inference. Communications Biology, 2024. Model: ESM2 + ResNet50 + Contact Map + Triplet Loss, [PDF], [Code], [Web Server].

  8. GloEC: A hierarchical-aware global model for predicting enzyme function. Briefings in Bioinformatics, 2024. Model: ESM-1b + Self Attention + GCN, [PDF], [Code].

  9. ifDEEPre: Large protein language-based deep learning enables interpretable and fast predictions of enzyme commission numbers Briefings in Bioinformatics, 2024. Model: ESM-1b + Self Attention + CNN, [PDF], [Code], [Web Server].

  10. FEDKEA: Enzyme function prediction with a large pretrained protein language model and distance-weighted k-nearest neighbor. bioRxiv, 2024. Model: ESM2 + MLP + KNN, [PDF], [Code].

  11. MAPred: Autoregressive Enzyme Function Prediction with Multi-scale Multi-modality Fusion. bioRxiv, 2024. Model: ESM2 + ProtTrans + CNN + Attention, [PDF].

  12. CLEAN: Enzyme function prediction using contrastive learning. Science, 2023. Model: ESM-1b + Triplet Loss, [PDF], [Code], [Web Server].

  13. DeepECtransformer: Functional annotation of enzyme-encoding genes using deep learning with transformer layers. Nature Communications, 2023. Model: ProtTrans + Self Attention + CNN, [PDF], [Code].

  14. HiFi-NN annotates the microbial dark matter with Enzyme Commission numbers NeurIPS, 2023. Model: ESM + MLP + Triplet Loss, [PDF].

  15. ECRECer: Enzyme Commission Number Prediction and Benchmarking with Hierarchical Dual-core Multitask Learning Framework. Research, 2023. Model: UniRep + ESM-1b + BiGRU + Attention, [PDF], [Code1], [Code2], [Web Server].

Deep Learning-Based Methods

  1. HIT-EC: Trust worthy prediction of enzyme commission numbers using a hierarchical interpretable transformer. bioRxiv, 2025. Model: Self Attention + Transformer, [PDF], [Code].

  2. TopEC: Prediction of Enzyme Commission classes by 3D graph neural networks and localized 3D protein descriptor. Nature Communications, 2025. Model: Structure descriptor + GNN, [PDF], [Code].

  3. ECPICK: Evidential deep learning for trustworthy prediction of enzyme commission number. Briefings in Bioinformatics, 2024. Model: Sequence encoding + CNN, [PDF], [Code], [Web Server].

  4. ProteInfer: Deep neural networks for protein functional inference. eLife, 2023. Model: Sequence encoding + CNN, [PDF], [Code].

  5. EnzBert: Predicting enzymatic function of protein sequences with attention. Bioinformatics, 2023. Model: Sequence encoding + Transformer, [PDF], [Code1], [Code2].

  6. deepNEC: A novel alignment-free tool for the identification and classification of nitrogen biochemical network-related enzymes using deep learning. Briefings in Bioinformatics, 2022. Model: Sequence encoding + CNN, [PDF], [Code], [Web Server].

  7. DAttProt: An Interpretable Double-Scale Attention Model for Enzyme Protein Class Prediction Based on Transformer Encoders and Multi-Scale Convolutions. Frontiers in Genetics, 2022. Model: Sequence encoding + Transformer + CNN, [PDF], [Code].

  8. DeepFRI: Structure-based protein function prediction using graph convolutional networks. Nature Communications, 2021. Model: LSTM + GCN + Contact Map, [PDF] [Web Server] [Code].

  9. HECNet: A hierarchical approach to enzyme function classification using a Siamese Triplet Network. Bioinformatics, 2020. Model: Sequence encoding + LSTM + CNN, [PDF] [Web Server].

  10. UDSMProt: Universal deep sequence models for protein classification. Bioinformatics, 2020. Model: Sequence encoding + LSTM , [PDF] [Code].

  11. DeepEC: Deep learning enables high-quality and highthroughput prediction of enzyme commission numbers. PNAS, 2019. Model: Sequence encoding + CNN, [PDF], [Code].

  12. mlDEEPre: Multi-Functional Enzyme Function Prediction With Hierarchical Multi-Label Deep Learning. Frontiers in Genetics, 2019. Model: Sequence encoding + CNN, [PDF].

  13. Prediction of Enzyme Function Based on Three Parallel Deep CNN and Amino Acid Mutation. International Journal of Molecular Sciences, 2019. Model: Sequence encoding + CNN, [PDF].

  14. DEEPre: Sequence-based enzyme EC number prediction by deep learning. Bioinformatics, 2018. Model: Sequence encoding + CNN, [PDF], [Web Server].

  15. EnzyNet: Enzyme classification using 3D convolutional neural networks on spatial representation. PeerJ, 2018. Model: Structure descriptor + CNN, [PDF], [Web Server].

Machine Learning-Based Methods

  1. PredictEFC: A fast and efficient multi‑label classifier for predicting enzyme family classes. BMC Bioinformatics, 2024. Model: SVM + RF, [PDF], [Web Server].

  2. The Classification of Enzymes by Deep Learning. IEEE Access, 2020. Model: SVM + KNN + LDA + MLP, [PDF],

  3. Alignment-Free Method to Predict Enzyme Classes and Subclasses. International Journal of Molecular Sciences, 2019. Model: LDA + MLP, [PDF],

  4. ECPred: A tool for the prediction of the enzymatic functions of protein sequences based on the EC nomenclature. BMC Bioinformatics, 2018. Model: SVM + KNN, [PDF], [Code], [Web Server].

  5. Automatic single- and multi-label enzymatic function prediction by machine learning. PeerJ, 2018. Model: SVM + MLP, [PDF], [Code].

  6. ML-KNN: Classifying Multifunctional Enzymes by Incorporating Three Different Models into Chou’s General Pseudo Amino Acid Composition. Journal of Membrane Biology, 2016. Model: KNN, [PDF].

  7. EFPrf: Prediction of Detailed Enzyme Functions and Identification of Specificity Determining Residues by Random Forests. PLoS One, 2013. Model: RF, [PDF].

  8. ECOH: An Enzyme Commission number predictor using mutual information and a support vector machine. Bioinformatics, 2013. Model: SVM, [PDF].

  9. Accurate prediction of protein enzymatic class by N-to-1 Neural Networks. BMC Bioinformatics, 2013. Model: N-to-1 Neural Network, [PDF].

  10. EFICAz 2.5: Application of a high-precision enzyme function predictor to 396 proteomes. Bioinformatics, 2012. Model: SVM , [PDF].

  11. Prediction of Enzyme Subfamily Class via Pseudo Amino Acid Composition by Incorporating the Conjoint Triad Feature. Protein & Peptide Letters, 2010. Model: SVM, [PDF].

  12. EFICAz 2: Enzyme function inference by a combined approach enhanced by machine learning. BMC Bioinformatics, 2009. Model: SVM, [PDF].

  13. EzyPred: A top–down approach for predicting enzyme functional classes and subclasses. Biochemical and Biophysical Research Communications , 2007. Model: KNN, [PDF].

Template Detection-Based Methods

  1. BENZ WS: The Bologna ENZyme Web Server for four-level EC number annotation. Nucleic Acids Research, 2021. Template: Sequence Similarity, [PDF], [Code].

  2. GrAPFI: Predicting enzymatic function of proteins from domain similarity graphs. BMC Bioinformatics, 2020. Template: Domain Similarity, [PDF], [Code].

  3. COFACTOR 2.0: Improved protein function prediction by combining structure, sequence and protein–protein interaction information. Nucleic Acids Research, 2017. Template: Structure Alignment, Sequence Alignment, PPI, [PDF], [Web Server].

  4. COFACTOR 1.0: An accurate comparative algorithm for structure-based protein function annotation. Nucleic Acids Research, 2012. Template: Structure Alignment, [PDF], [Web Server].

  5. EnzML: Multi-label prediction of enzyme classes using InterPro signatures. BMC Bioinformatics, 2012. Template: Sequence Alignment, [PDF].

  6. ModEnzA: Accurate Identification ofMetabolic Enzymes Using Function Specific Profile HMMs with Optimised Discrimination Threshold andModified Emission Probabilities. Advances in Bioinformatics, 2011. Template: Sequence Similarity, [PDF].

  7. EAT Server: Evolutionary Trace Annotation Server: automated enzyme function prediction in protein structures using 3D templates. Bioinformatics, 2009. Template: Structure Alignment, [PDF], [Web Server].

  8. EAT: Prediction of enzyme function based on 3D templates of evolutionarily important amino acids. BMC Bioinformatics, 2009. Template: Structure Alignment, [PDF].

  9. 3D-Fun: Predicting enzyme function from structure Nucleic Acids Research, 2008. Template: Structure Alignment, [PDF].

  10. EFICAz: A comprehensive approach for accurate genome-scale enzyme function inference. Nucleic Acids Research, 2004. Template: Sequence Similarity, [PDF].

  11. PRIAM: Enzyme‐specific profiles for genome annotation. Nucleic Acids Research, 2003. Template: Sequence Similarity, [PDF].