Few-Shot Bioactivity Prediction with Meta-Learning under Assay Heterogeneity

2026-10-05

AI Summary

This study addresses the low reliability of bioactivity prediction in early drug discovery caused by assay data scarcity and heterogeneity. To this end, it proposes MetaHeta, a meta-learning framework that introduces a novel linear-exact hybrid attention architecture. Specifically, the method leverages linear attention over large auxiliary datasets to correlate heterogeneous data, while employing exact attention on scarce task contexts to mitigate assay interference. Evaluated on the ChEMBL and BindingDB benchmarks, the proposed framework significantly improves few-shot bioactivity prediction accuracy. Furthermore, it effectively enhances compound prioritization performance in retrospective Bayesian optimization campaigns.

Abstract

Accurate bioactivity prediction is a central challenge in early-stage drug discovery, as individual assays often contain too few measurements to train reliable models independently. Meta-learning offers a principled approach to this few-shot setting, but assay heterogeneity may limit its effectiveness. Here, we test this hypothesis and show that meta-learning performance degrades as meta-training tasks become more heterogeneous. To address this, we introduce MetaHeta, a meta-learning framework that accounts for assay heterogeneity by conditioning predictions on auxiliary data from related assays, with relatedness defined flexibly from available assay information. The architecture of MetaHeta combines linear attention over large auxiliary datasets with exact attention over scarce task-specific context, enabling efficient scaling to the former without compromising exact attention over the latter. We demonstrate the benefits of our approach on assays from ChEMBL and BindingDB, improving few-shot bioactivity prediction and downstream compound prioritization in retrospective Bayesian optimization.

Problem

• few-shot bioactivity prediction
• meta-learning
• assay heterogeneity
• drug discovery

Innovation

• Meta-Learning
• Assay Heterogeneity
• Few-Shot Bioactivity Prediction
• Linear Attention
• MetaHeta