PT  - JOURNAL ARTICLE
AU  - Inecik, Kemal
AU  - Uhlmann, Andreas
AU  - Lotfollahi, Mohammad
AU  - Theis, Fabian
TI  - MultiCPA: Multimodal Compositional Perturbation Autoencoder
AID  - 10.1101/2022.07.08.499049
DP  - 2022 Jan 01
TA  - bioRxiv
PG  - 2022.07.08.499049
4099  - http://biorxiv.org/content/early/2022/07/10/2022.07.08.499049.short
4100  - http://biorxiv.org/content/early/2022/07/10/2022.07.08.499049.full
AB  - Single-cell multimodal profiling provides a high-resolution view of cellular information. Recently, multimodal profiling approaches have been coupled with CRISPR technologies to perform pooled screens of single or combinatorial perturbations. This opens the possibility of exploring the massive space of combinatorial perturbations and their regulatory effects computationally from the extrapolation of a few experimentally feasible combinations. Here, we propose MultiCPA, an end-to-end generative architecture to predict multimodal perturbation response at single cell level. Two mixing strategies to integrate multiple modalities are introduced and compared with existing methods. MultiCPA was also shown to accurately predict unseen combinatorial perturbation responses for multiple modalities. The code to reproduce the results is available on GitHub, theislab/multicpa.Competing Interest StatementF.J.T. consults for Immunai Inc., Singularity Bio B.V., CytoReason Ltd, and Omniscope Ltd, and has ownership interest in Dermagnostix GmbH and Cellarity.