Skip to main content

Revealing and avoiding bias in semantic similarity scores for protein pairs