DeepTM: A deep learning algorithm for prediction of melting temperature of thermophilic proteins directly from sequences

Mengyu Li; Hongzhao Wang; Zhenwu Yang; Longgui Zhang; Yushan Zhu

doi:10.1016/j.csbj.2023.11.006

DeepTM: A deep learning algorithm for prediction of melting temperature of thermophilic proteins directly from sequences

Comput Struct Biotechnol J. 2023 Nov 4:21:5544-5560. doi: 10.1016/j.csbj.2023.11.006. eCollection 2023.

Authors

Mengyu Li¹, Hongzhao Wang¹, Zhenwu Yang¹, Longgui Zhang², Yushan Zhu^{1

3}

Affiliations

¹ College of Life Science and Technology, Beijing University of Chemical Technology, Beijing 100029, China.
² SINOPEC Beijing Research Institute of Chemical Industry, Beijing 100013, China.
³ National Energy R&D Center for Biorefinery, Beijing University of Chemical Technology, Beijing 100029, China.

Abstract

Thermally stable proteins find extensive applications in industrial production, pharmaceutical development, and serve as a highly evolved starting point in protein engineering. The thermal stability of proteins is commonly characterized by their melting temperature (T_m). However, due to the limited availability of experimentally determined T_m data and the insufficient accuracy of existing computational methods in predicting T_m, there is an urgent need for a computational approach to accurately forecast the T_m values of thermophilic proteins. Here, we present a deep learning-based model, called DeepTM, which exclusively utilizes protein sequences as input and accurately predicts the T_m values of target thermophilic proteins on a dataset consisting of 7790 thermophilic protein entries. On a test set of 1550 samples, DeepTM demonstrates excellent performance with a coefficient of determination (R²) of 0.75, Pearson correlation coefficient (P) of 0.87, and root mean square error (RMSE) of 6.24 ℃. We further analyzed the sequence features that determine the thermal stability of thermophilic proteins and found that dipeptide frequency, optimal growth temperature (OGT) of the host organisms, and the evolutionary information of the protein significantly affect its melting temperature. We compared the performance of DeepTM with recently reported methods, ProTstab2 and DeepSTABp, in predicting the T_m values on two blind test datasets. One dataset comprised 22 PET plastic-degrading enzymes, while the other included 29 thermally stable proteins of broader classification. In the PET plastic-degrading enzyme dataset, DeepTM achieved RMSE of 8.25 ℃. Compared to ProTstab2 (20.05 ℃) and DeepSTABp (20.97 ℃), DeepTM demonstrated a reduction in RMSE of 58.85% and 60.66%, respectively. In the dataset of thermally stable proteins, DeepTM (RMSE=7.66 ℃) demonstrated a 51.73% reduction in RMSE compared to ProTstab2 (RMSE=15.87 ℃). DeepTM, with the sole requirement of protein sequence information, accurately predicts the melting temperature and achieves a fully end-to-end prediction process, thus providing enhanced convenience and expediency for further protein engineering.

Keywords: Deep learning predictor; Graph convolutional neural network; Melting temperature; Protein thermal stability; Self-attention network; Sequence features.