Delighted to be involved in this recent preprint led by our friends at @cvrinfo.bsky.social @cvrbioinfo.bsky.social @davidlrobertson.bsky.social Protein language models learn underlying mutation biases alongside fitness landscapes www.biorxiv.org/content/10.6...
Protein language models learn underlying mutation biases alongside fitness landscapesProtein language models (PLMs) score the effects of amino acid replacements as pseudo-probabilities, which are widely utilised to map protein fitness landscapes. However, because their training data r...www.biorxiv.org
