Likelihood ratio tests for a large directed acyclic graph

Nonlinear causal discovery with confounders

This article introduces a causal discovery method to learn nonlinear relationships in a directed acyclic graph with correlated Gaussian errors due to confounding. First, we derive model identifiability under the sublinear growth assumption. Then, we propose a novel method, named the Deconfounded Functional Structure Estimation (DeFuSE), consisting of a deconfounding adjustment to remove the confounding effects and a sequential procedure to estimate the causal order of variables. We implement DeFuSE via feedforward neural networks for scalable computation. Moreover, we establish the consistency of DeFuSE under an assumption called the strong causal minimality. In simulations, DeFuSE compares favorably against state-of-the-art competitors that ignore confounding or nonlinearity. Finally, we demonstrate the utility and effectiveness of the proposed approach with an application to gene regulatory network analysis. The Python implementation is available at https://github.com/chunlinli/defuse.

Inference for a Large Directed Acyclic Graph with Unspecified Interventions

Statistical inference of directed relations given some unspecified interventions (i.e., the intervention targets are unknown) is challenging. In this article, we test hypothesized directed relations with unspecified interventions. First, we derive conditions to yield an identifiable model. Unlike classical inference, testing directed relations requires to identify the ancestors and relevant interventions of hypothesis-specific primary variables. To this end, we propose a peeling algorithm based on nodewise regressions to establish a topological order of primary variables. Moreover, we prove that the peeling algorithm yields a consistent estimator in low-order polynomial time. Second, we propose a likelihood ratio test integrated with a data perturbation scheme to account for the uncertainty of identifying ancestors and interventions. Also, we show that the distribution of a data perturbation test statistic converges to the target distribution. Numerical examples demonstrate the utility and effectiveness of the proposed methods, including an application to infer gene regulatory networks. The R implementation is available at https://github.com/chunlinli/intdag.

Functional structural equation model

In this article, we introduce a functional structural equation model for estimating directional relations from multivariate functional data. We decouple the estimation into two major steps: directional order determination and selection through sparse functional regression. We first propose a score function at the linear operator level, and show that its minimization can recover the true directional order when the relation between each function and its parental functions is nonlinear. We then develop a sparse functional additive regression, where both the response and the multivariate predictors are functions and the regression relation is additive and nonlinear. We also propose strategies to speed up the computation and scale up our method. In theory, we establish the consistencies of order determination, sparse functional additive regression, and directed acyclic graph estimation, while allowing both the dimension of the Karhunen-Loéve expansion coefficients and the number of random functions to diverge with the sample size. We illustrate the efficacy of our method through simulations, and an application to brain effective connectivity analysis.

Penalized mediation models for multivariate data

Statistical methods to integrate multiple layers of data, from exposures to intermediate traits to outcome variables, are needed to guide interpretation of complex data sets for which variables are likely contributing in a causal pathway from exposure to outcome. Statistical mediation analysis based on structural equation models provide a general modeling framework, yet they can be difficult to apply to high-dimensional data and they are not automated to select the best fitting model. To overcome these limitations, we developed novel algorithms and software to simultaneously evaluate multiple exposure variables, multiple intermediate traits, and multiple outcome variables. Our penalized mediation models are computationally efficient and simulations demonstrate that they produce reliable results for large data sets. Application of our methods to a study of vascular disease demonstrates their utility to identify novel direct effects of single-nucleotide polymorphisms (SNPs) on coronary heart disease and peripheral artery disease, while disentangling the effects of SNPs on the intermediate risk factors including lipids, cigarette smoking, systolic blood pressure, and type 2 diabetes.

Corrected score methods for estimating Bayesian networks with error-prone nodes

Motivated by inferring cellular signaling networks using noisy flow cytometry data, we develop procedures to draw inference for Bayesian networks based on error-prone data. Two methods for inferring causal relationships between nodes in a network are proposed based on penalized estimation methods that account for measurement error and encourage sparsity. We discuss consistency of the proposed network estimators and develop an approach for selecting the tuning parameter in the penalized estimation methods. Empirical studies are carried out to compare the proposed methods with a naive method that ignores measurement error. Finally, we apply these methods to infer signaling networks using single cell flow cytometry data.

Inferring causal direction between two traits in the presence of horizontal pleiotropy with GWAS summary data

Orienting the causal relationship between pairs of traits is a fundamental task in scientific research with significant implications in practice, such as in prioritizing molecular targets and modifiable risk factors for developing therapeutic and interventional strategies for complex diseases. A recent method, called Steiger's method, using a single SNP as an instrument variable (IV) in the framework of Mendelian randomization (MR), has since been widely applied. We report the following new contributions. First, we propose a single SNP-based alternative, overcoming a severe limitation of Steiger's method in simply assuming, instead of inferring, the existence of a causal relationship. We also clarify a condition necessary for the validity of the methods in the presence of hidden confounding. Second, to improve statistical power, we propose combining the results from multiple, and possibly correlated, SNPs as multiple instruments. Third, we develop three goodness-of-fit tests to check modeling assumptions, including those required for valid IVs. Fourth, by relaxing one of the three IV assumptions in MR, we propose several methods, including an Egger regression-like approach and its multivariable version (analogous to multivariable MR), to account for horizontal pleiotropy of the SNPs/IVs, which is often unavoidable in practice. All our methods can simultaneously infer both the existence and (if so) the direction of a causal relationship, largely expanding their applicability over that of Steiger's method. Although we focus on uni-directional causal relationships, we also briefly discuss an extension to bi-directional relationships. Through extensive simulations and an application to infer the causal directions between low density lipoprotein (LDL) cholesterol, or high density lipoprotein (HDL) cholesterol, and coronary artery disease (CAD), we demonstrate the superior performance and advantage of our proposed methods over Steiger's method and bi-directional MR. In particular, after accounting for horizontal pleiotropy, our method confirmed the well known causal direction from LDL to CAD, while other methods, including bi-directional MR, might fail.

A Regularization-Based Adaptive Test for High-Dimensional Generalized Linear Models

In spite of its urgent importance in the era of big data, testing high-dimensional parameters in generalized linear models (GLMs) in the presence of high-dimensional nuisance parameters has been largely under-studied, especially with regard to constructing powerful tests for general (and unknown) alternatives. Most existing tests are powerful only against certain alternatives and may yield incorrect Type I error rates under high-dimensional nuisance parameter situations. In this paper, we propose the adaptive interaction sum of powered score (aiSPU) test in the framework of penalized regression with a non-convex penalty, called truncated Lasso penalty (TLP), which can maintain correct Type I error rates while yielding high statistical power across a wide range of alternatives. To calculate its p-values analytically, we derive its asymptotic null distribution. Via simulations, its superior finite-sample performance is demonstrated over several representative existing methods. In addition, we apply it and other representative tests to an Alzheimer's Disease Neuroimaging Initiative (ADNI) data set, detecting possible gene-gender interactions for Alzheimer's disease. We also put R package "aispu" implementing the proposed test on GitHub.