Interpretable Enzyme Function Prediction via Residue-Level Detection
This work addresses the problem of interpretable enzyme function prediction for bioinformatics, offering a novel approach to detect function-specific local residues, though it is incremental in applying detection methods to this domain.
The paper tackles the challenge of predicting multiple enzyme functions from sequences, which is a sparse multi-label classification problem, by introducing ProtDETR, an attention-based framework that casts it as a detection problem and significantly outperforms existing deep learning methods.
Predicting multiple functions labeled with Enzyme Commission (EC) numbers from the enzyme sequence is of great significance but remains a challenge due to its sparse multi-label classification nature, i.e., each enzyme is typically associated with only a few labels out of more than 6000 possible EC numbers. However, existing machine learning algorithms generally learn a fixed global representation for each enzyme to classify all functions, thereby they lack interpretability and the fine-grained information of some function-specific local residue fragments may be overwhelmed. Here we present an attention-based framework, namely ProtDETR (Protein Detection Transformer), by casting enzyme function prediction as a detection problem. It uses a set of learnable functional queries to adaptatively extract different local representations from the sequence of residue-level features for predicting different EC numbers. ProtDETR not only significantly outperforms existing deep learning-based enzyme function prediction methods, but also provides a new interpretable perspective on automatically detecting different local regions for identifying different functions through cross-attentions between queries and residue-level features. Code is available at https://github.com/yangzhao1230/ProtDETR.