Seq2Topt: a sequence-based deep learning predictor of enzyme optimal temperature

An accurate deep learning predictor is needed for enzyme optimal temperature (Topt), which quantitatively describes how temperature affects the enzyme catalytic activity. Seq2Topt, developed in this study, reached a superior accuracy on Topt prediction just using protein sequences (RMSE = 13.3 and R2=0.48) in comparison with existing models, and could capture key protein regions for enzyme Topt with multi-head attention on residues. Through case studies on thermophilic enzyme selection and predicting enzyme Topt shifts caused by point mutations, Seq2Topt was demonstrated as a promising computational tool for enzyme mining and in-silico enzyme design. Additionally, accurate deep learning predictors of enzyme optimal pH (Seq2pHopt, RMSE=0.92 and R2=0.37) and melting temperature (Seq2Tm, RMSE=7.57 and R2=0.64) were developed based on the model architecture of Seq2Topt, suggesting that the development of Seq2Topt could potentially give rise to a useful prediction platform of enzymes.