Abstract:Understanding protein thermostability is essential for various biotechnological and biological applications. However, traditional experimental methods for assessing this property are time-consuming, expensive, and error-prone. Recently, the application of Deep Learning techniques from Natural Language Processing (NLP) was extended to the field of biology, with an emphasis on protein modeling. From a linguistic perspective, the primary sequence of proteins can be viewed as a string of amino acids that follow a physicochemical grammar. This study explores the potential of Deep Learning models trained on protein sequences to predict protein thermostability which provide improvements with respect to current approaches. We implemented TemBERTure, a Deep Learning framework to classify the thermal class (non-thermophilic or thermophilic) and predict and melting temperature of a protein, based on its primary sequence. Our findings highlight the critical role that data diversity plays on training robust models. Models trained on datasets with a wider range of sequences from various organisms exhibited superior performance compared to those with limited diversity. This emphasizes the need for a comprehensive data curation strategy that ensures a balanced representation of diverse species in the training data, to avoid the risk that the model focuses on recognizing the evolutionary lineage of the sequence rather than the intrinsic thermostability features. In order to gain more nuanced insights into protein thermostability, we propose leveraging attention scores within Deep Learning models to gain more nuanced insights into protein thermostability. We show that analyzing these scores alongside the 3D protein structure could offer a better understanding of the complex interplay between amino acid properties, their positioning, and the surrounding microenvironment, all crucial factors influencing protein thermostability. This work sheds light on the limitations of current protein thermostability prediction methods and introduces new avenues for exploration. By emphasizing data diversity and utilizing refined attention scores, future research can pave the way for more accurate and informative methods for predicting protein thermostability.

Learning deep representations of enzyme thermal adaptation

TemBERTure: Advancing protein thermostability prediction with Deep Learning and attention mechanisms

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

Thermal Adaptation of Cytosolic Malate Dehydrogenase Revealed by Deep Learning and Coevolutionary Analysis

Improving enzyme optimum temperature prediction with resampling strategies and ensemble learning

Data-driven strategies for the computational design of enzyme thermal stability: trends, perspectives, and prospects

Enhancing Machine-Learning Prediction of Enzyme Catalytic Temperature Optima through Amino Acid Conservation Analysis

A learnable transition from low temperature to high temperature proteins with neural machine translation

DLTKcat: deep learning-based prediction of temperature-dependent enzyme turnover rates

Adaptations of protein structure and function to temperature: there is more than one way to ‘skin a cat’

Evolutionary context-integrated deep sequence modeling for protein engineering

DeepEnzyme: a robust deep learning model for improved enzyme turnover number prediction by utilizing features of protein 3D-structures

Parallel molecular mechanisms for enzyme temperature adaptation

Enzyme evolution and the temperature dependence of enzyme catalysis

DeepTP: A Deep Learning Model for Thermophilic Protein Prediction

Accurate computational evolution of proteins and its dependence on deep learning

Semantical and Geometrical Protein Encoding Toward Enhanced Bioactivity and Thermostability

ECNet is an evolutionary context-integrated deep learning framework for protein engineering

Deep Learning Basedkcatprediction Enables Improved Enzyme Constrained Model Reconstruction

Prediction of Thermostability of Enzymes Based on the Amino Acid Index (AAindex) Database and Machine Learning

Scalable emulation of protein equilibrium ensembles with generative deep learning