AI challenges for predicting the impact of mutations on protein stability

TL;DR

This paper reviews AI-based methods for predicting how mutations affect protein stability, highlighting their limitations, biases, and the stagnation in accuracy over 15 years, and discusses future challenges.

Contribution

It provides a critical analysis of existing AI predictors for protein stability, emphasizing their limitations and the need for new approaches to improve performance.

Findings

Predictor accuracy has stagnated at around 1 kcal/mol for over 15 years.

Current methods exhibit biases towards training data and limited generalizability.

There is a need to address interpretability and develop more robust models.

Abstract

Stability is a key ingredient of protein fitness and its modification through targeted mutations has applications in various fields such as protein engineering, drug design and deleterious variant interpretation. Many studies have been devoted over the past decades to building new, more effective methods for predicting the impact of mutations on protein stability, based on the latest developments in artificial intelligence (AI). We discuss their features, algorithms, computational efficiency, and accuracy estimated on an independent test set. We focus on a critical analysis of their limitations, the recurrent biases towards the training set, their generalizability and interpretability. We found that the accuracy of the predictors has stagnated at around 1 kcal/mol for over 15 years. We conclude by discussing the challenges that need to be addressed to reach improved performance.