Neural scaling laws for phenotypic drug discovery

TL;DR

This paper explores how scaling up neural networks and data affects performance in phenotypic drug discovery, introducing a novel pretraining task that improves scalability and accuracy on drug development benchmarks.

Contribution

The study introduces the Inverse Biological Process pretraining task, which enables neural networks to better scale and perform on drug discovery tasks compared to traditional supervised methods.

Findings

DNNs do not improve with scale when trained directly on drug discovery tasks.

Pretraining with IBP significantly enhances DNN performance.

Performance of IBP-trained DNNs improves monotonically with data and model size.

Abstract

Recent breakthroughs by deep neural networks (DNNs) in natural language processing (NLP) and computer vision have been driven by a scale-up of models and data rather than the discovery of novel computing paradigms. Here, we investigate if scale can have a similar impact for models designed to aid small molecule drug discovery. We address this question through a large-scale and systematic analysis of how DNN size, data diet, and learning routines interact to impact accuracy on our Phenotypic Chemistry Arena (Pheno-CA) benchmark: a diverse set of drug development tasks posed on image-based high content screening data. Surprisingly, we find that DNNs explicitly supervised to solve tasks in the Pheno-CA do not continuously improve as their data and model size is scaled-up. To address this issue, we introduce a novel precursor task, the Inverse Biological Process (IBP), which is designed to resemble the causal objective functions that have proven successful for NLP. We indeed find that DNNs first trained with IBP then probed for performance on the Pheno-CA significantly outperform task-supervised DNNs. More importantly, the performance of these IBP-trained DNNs monotonically improves with data and model scale. Our findings reveal that the DNN ingredients needed to accurately solve small molecule drug development tasks are already in our hands, and project how much more experimental data is needed to achieve any desired level of improvement. We release our Pheno-CA benchmark and code to encourage further study of neural scaling laws for small molecule drug discovery.