Enhancing Coevolutionary Signals In Protein-protein Interaction Prediction Through Clade-wise Alignment Integration

This article discusses the use of deep learning methods, specifically AlphaFold, in predicting protein interactions. However, there are limitations to its use in genome-wide searches due to computational resources and training data. The authors propose a pipeline that can be used as a pre-filter step to identify potential interacting protein pairs for further analysis.