Back

Minimal epistatic networks from integrated sequence and mutational protein data.

Cocco, S.; Monasson, R.; Posani, L.

2023-09-25 bioinformatics
10.1101/2023.09.25.559251 bioRxiv
Show abstract

Predicting the functional effects of mutations to a wild-type protein sequence is a major computational challenge. We introduce here a computationally efficient procedure to identify the few, most informative epistatic links between residues in a protein, integrating sequence data and functional measurements with mutational scans. Our approach shows performances comparable to state-of-the-art deep networks, while requiring much less parameters and being hence much more interpretable. The selected network links mostly focus on the protein functional sites, adapt to the in vitro or in vivo function experimentally tested, and are not necessary related to structural contacts.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.