Predicting Cellular Responses to Novel Drug Perturbations at a Single-Cell Resolution

TL;DR

This paper presents chemCPA, a novel model that leverages bulk RNA data to predict cellular responses to new drugs at single-cell resolution, reducing experimental costs and accelerating drug discovery.

Contribution

Introduction of chemCPA, an encoder-decoder model with transfer learning architecture to predict drug effects on cells using bulk RNA data.

Findings

Improved generalization in predicting drug responses.

Reduced need for costly single-cell experiments.

Facilitated in-silico hypothesis generation.

Abstract

Single-cell transcriptomics enabled the study of cellular heterogeneity in response to perturbations at the resolution of individual cells. However, scaling high-throughput screens (HTSs) to measure cellular responses for many drugs remains a challenge due to technical limitations and, more importantly, the cost of such multiplexed experiments. Thus, transferring information from routinely performed bulk RNA HTS is required to enrich single-cell data meaningfully. We introduce chemCPA, a new encoder-decoder architecture to study the perturbational effects of unseen drugs. We combine the model with an architecture surgery for transfer learning and demonstrate how training on existing bulk RNA HTS datasets can improve generalisation performance. Better generalisation reduces the need for extensive and costly screens at single-cell resolution. We envision that our proposed method will facilitate more efficient experiment designs through its ability to generate in-silico hypotheses, ultimately accelerating drug discovery.