Multi-Scale Representation Learning for Protein Fitness Prediction

Given the limited availability of functional annotations from wet-lab experiments, previous methods have primarily relied on self-supervised models trained on vast, unlabeled protein sequence or structure datasets.