Sequence-based coevolutionary prediction of species-specific interactomes.

ContextMirror2.0 predicts species-specific protein-protein interactions for Escherichia coli, Salmonella enterica, and Staphylococcus aureus, with about 40% of its top-1000 predictions supported by experimental PPI databases.

• Why it matters: Predicting protein function from conserved PPIs is incomplete, and computational PPI methods are limited by sparse functional annotations, especially for species-specific interaction networks.
• What they did: The study developed ContextMirror2.0, a coevolution-based approach that avoids labelled training data by using phylogenetic coevolution signals to generate predicted interactomes.
• The result: Functional community comparisons show both highly conserved interaction patterns and species-specific interactions, enabling hypotheses about how species variations may shape functional outcomes and adaptation.

AI summary of the abstract · may contain errors · not medical advice · read the paper.