Directed Graphical Models and Causal Discovery for Zero-Inflated Data

Shiqing Yu, Mathias Drton, Ali Shojaie

Keywords: Bayesian network, causal discovery, directed acyclic graph, identifiability

TL;DR: The paper establishes identifiability of causal effects from zero-inflated data and develops estimation procedures for learning DAGs.

Abstract: With advances in technology, gene expression measurements from single cells can be used to gain refined insights into regulatory relationships among genes. Directed graphical models are well-suited to explore such (cause-effect) relationships. However, statistical analyses of single cell data are complicated by the fact that the data often show zero-inflated expression patterns. To address this challenge, we propose directed graphical models that are based on Hurdle conditional distributions parametrized in terms of polynomials in parent variables and their $0/1$ indicators of being zero or nonzero. While directed graphs for Gaussian models are only identifiable up to an equivalence class in general, we show that, under a natural and weak assumption, the exact directed acyclic graph of our zero-inflated models can be identified. We propose methods for graph recovery, apply our model to real single-cell gene expression data on T helper cells, and show simulated experiments that validate the identifiability and graph estimation methods in practice.