Target-Free Compound Activity Prediction via Few-Shot Learning
This work addresses a key limitation in drug discovery by enabling more nuanced activity predictions, though it is incremental in extending few-shot learning from binary to continuous outputs.
The paper tackles the problem of predicting continuous compound activities in target-free drug discovery using few-shot learning, achieving superior performance over existing methods on various datasets.
Predicting the activities of compounds against protein-based or phenotypic assays using only a few known compounds and their activities is a common task in target-free drug discovery. Existing few-shot learning approaches are limited to predicting binary labels (active/inactive). However, in real-world drug discovery, degrees of compound activity are highly relevant. We study Few-Shot Compound Activity Prediction (FS-CAP) and design a novel neural architecture to meta-learn continuous compound activities across large bioactivity datasets. Our model aggregates encodings generated from the known compounds and their activities to capture assay information. We also introduce a separate encoder for the unknown compound. We show that FS-CAP surpasses traditional similarity-based techniques as well as other state of the art few-shot learning methods on a variety of target-free drug discovery settings and datasets.