AF2-Mutation: Adversarial Sequence Mutations against AlphaFold2 on Protein Tertiary Structure Prediction

TL;DR

This paper investigates the robustness of AlphaFold2 against adversarial sequence mutations, demonstrating that small sequence changes can significantly alter structure predictions and identifying key residues affecting protein conformation.

Contribution

It introduces an evolutionary approach to generate adversarial mutations that significantly impact AlphaFold2's predictions and identifies critical residues influencing protein structure.

Findings

Modifying three residues can change AF2's predictions by 46.61 in lDDT score.

The method identifies biologically meaningful residues critical for structure.

Adversarial mutations can suggest alternative protein conformations.

Abstract

Deep learning-based approaches, such as AlphaFold2 (AF2), have significantly advanced protein tertiary structure prediction, achieving results comparable to real biological experimental methods. While AF2 has shown limitations in predicting the effects of mutations, its robustness against sequence mutations remains to be determined. Starting with the wild-type (WT) sequence, we investigate adversarial sequences generated via an evolutionary approach, which AF2 predicts to be substantially different from WT. Our experiments on CASP14 reveal that by modifying merely three residues in the protein sequence using a combination of replacement, deletion, and insertion strategies, the alteration in AF2's predictions, as measured by the Local Distance Difference Test (lDDT), reaches 46.61. Moreover, when applied to a specific protein, SPNS2, our proposed algorithm successfully identifies biologically meaningful residues critical to protein structure determination and potentially indicates alternative conformations, thus significantly expediting the experimental process.