Discovering Root Causal Genes with High Throughput Perturbations

Abstract

The root causes of disease correspond to the initial perturbations to a biological system that generate patient symptoms as a downstream effect. Identifying root causal genes is critical towards developing treatments that modify disease at its onset. However, RNA-sequencing (RNA-seq) data introduces a host of challenges such as measurement error, high dimensionality and non-linearity that compromise accurate estimation of root causal effects even with state-of-the-art methods. We therefore instead leverage Perturb-seq, or high throughput single cell perturbations, to learn the causal order between the genes. We then transfer the causal order to bulk RNA-seq and identify root causal genes specific to a given patient using a novel statistic quantifying the strength of a type of root causal effect. Experiments demonstrate large improvements relative to existing algorithms. Applications to macular degeneration and multiple sclerosis also reveal root causal genes that lie on known pathogenic pathways, delineate patient subgroups and implicate a newly defined omnigenic root causal model. Code is available at github.com/ericstrobl/RCSP.