Massively multiplex multi-modal chemical screens at single-cell resolution

We present icCITE-plex and DOGMA-plex, two novel cellular barcoding systems that enable multiplexed analysis of multiple independent samples within a single experiment through a combination of viral barcodes and antibody-based plate-barcodes. Using this platform, we successfully profiled 360 drug-dose responses across 475 surface proteins, 64 intracellular epitopes, and the whole transcriptome in human and mouse T cells treated with 56 different kinase inhibitors, demonstrating the method's ability to capture multi-modal cellular responses to chemical perturbations. To analyze this complex dataset, we developed MoCAVI, a new computational method that effectively separates perturbation effects from background cell heterogeneity while integrating multiple data modalities, outperforming existing methods in identifying drug mechanisms of action.