Neural Potts Model

Abstract

We propose the Neural Potts Model objective as an amortized optimization problem. The objective enables training a single model with shared parameters to explicitly model energy landscapes across multiple protein families. Given a protein sequence as input, the model is trained to predict a pairwise coupling matrix for a Potts model energy function describing the local evolutionary landscape of the sequence. Plausible couplings are predicted for train and validation sequences. A controlled ablation experiment assessing unsupervised contact prediction on sets of related protein families finds a gain from amortization for low-depth MSAs; the result is confirmed on a larger database with broad coverage of protein sequences.

Tasks

Reproductions