Mixed graphical models for integrative causal analysis with application to chronic lung disease diagnosis and prognosis

Sedgewick, Andrew J.; Buschur, Kristina L.; Shi, Ivy; Ramsey, Joseph D.; Raghu, Vineet K.; Manatakis, Dimitris V.; Zhang, Yingze; Bon, Jessica; Chandra, Divay; Karoleski, Chad; Sciurba, Frank C.; Spirtes, Peter; Glymour, Clark; Benos, Panayiotis V.

doi:10.1093/bioinformatics/bty769

Cited by 56 publications

(59 citation statements)

References 40 publications

(51 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This feature is particularly useful when the study aims to address simultaneous causal relationships in "omics"scale data sets, for example in studies of gene expression [41] or metabolites [42]. Recent network-building methods have been developed that allow the analysis of potentially hundreds of variables, including both discrete and continuous data types, taking advantage of the ability of genetic variables to operate as causal anchors to help orient the direction of relationships between non-genetic variables [43][44][45][46][47]. These BN approaches could be considered complementary to the MR-based approaches [28] that enable the construction of such networks, as they use very different algorithms, are more restrictive in terms of requiring individual level input data, and produce different outputs, albeit in order to achieve similar goals.…”

Section: Plos Geneticsmentioning

confidence: 99%

Bayesian network analysis incorporating genetic anchors complements conventional Mendelian randomization approaches for exploratory analysis of causal relationships in complex data

et al. 2020

View full text Add to dashboard Cite

Mendelian randomization (MR) implemented through instrumental variables analysis is an increasingly popular causal inference tool used in genetic epidemiology. But it can have limitations for evaluating simultaneous causal relationships in complex data sets that include, for example, multiple genetic predictors and multiple potential risk factors associated with the same genetic variant. Here we use real and simulated data to investigate Bayesian network analysis (BN) with the incorporation of directed arcs, representing genetic anchors, as an alternative approach. A Bayesian network describes the conditional dependencies/independencies of variables using a graphical model (a directed acyclic graph) with an accompanying joint probability. In real data, we found BN could be used to infer simultaneous causal relationships that confirmed the individual causal relationships suggested by bi-directional MR, while allowing for the existence of potential horizontal pleiotropy (that would violate MR assumptions). In simulated data, BN with two directional anchors (mimicking genetic instruments) had greater power for a fixed type 1 error than bidirectional MR, while BN with a single directional anchor performed better than or as well as bi-directional MR. Both BN and MR could be adversely affected by violations of their underlying assumptions (such as genetic confounding due to unmeasured horizontal pleiotropy). BN with no directional anchor generated inference that was no better than by chance, emphasizing the importance of directional anchors in BN (as in MR). Under highly pleiotropic simulated scenarios, BN outperformed both MR (and its recent extensions) and two recently-proposed alternative approaches: a multi-SNP mediation intersection-union test (SMUT) and a latent causal variable (LCV) test. We conclude that BN incorporating genetic anchors is a useful complementary method to conventional MR for exploring PLOS GENETICS

show abstract

Section: Plos Geneticsmentioning

confidence: 99%

Bayesian network analysis incorporating genetic anchors complements conventional Mendelian randomization approaches for exploratory analysis of causal relationships in complex data

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Bayesian Networks (BNs)based dependency modeling is an established computational biology tool that has been rapidly gaining acceptance in big biological data analysis (Branciamore et al 2018;Cooper et al 2015;Gogoshin, Boerwinkle, and Rodin 2017;Jiang, Barmada, and Visweswaran. 2010;Lan et al 2016;Neapolitan, Xue, and Jiang 2014;Needham et al 2007;Pe'er 2005;Piatetsky-Shapiro and Tamayo 2003;Qi, Li, and Cheng 2014;Rodin et al 2005;Rodin et al 2012;Sedgewick et al 2019;Sherif, Zayed, and Fakhr 2015;Wang, Audenaert, and Michoel 2019;Yin et al 2015;Zeng, Jiang, and Neapolitan 2016;Ziebarth, Bhattacharya, and Cui 2013;Zhang and Shi 2017;Zhang et al 2017;Zhang et al 2014). Comprehensive treatments of BN methodology, and probabilistic networks (PNs) in general, can be found in numerous textbooks (Pearl 1988;Pearl 2009;Russell and Norvig.…”

Section: Introductionmentioning

confidence: 99%

Synthetic data generation with probabilistic Bayesian Networks

Gogoshin

Branciamore

Rodin

2020

Preprint

View full text Add to dashboard Cite

Bayesian Network (BN) modeling is a prominent and increasingly popular computational systems biology method.It aims at constructing probabilistic networks from the big heterogeneous biological data that reflect the underlying networks of biological relationships. Currently, there is a variety of strategies for evaluating the BN methodology performance, ranging from utilizing the artificial benchmark datasets and models, to the specialized biological benchmark datasets, to the simulation studies that generate synthetic data from the predefined network models. The latter is arguably the most comprehensive approach; however, existing implementations are typically limited by their reliance on the SEM (structural equation modeling) framework, with many explicit and implicit assumptions that might not be realistic in a typical biological data analysis scenario. In this study, we develop an alternative, purely probabilistic, simulation framework that is a more appropriate fit with the real biological data and biological network models. In conjunction, we also expand on our current understanding of the theoretical notions of causality and dependence / conditional independence in BNs and Markov Blankets within.

show abstract

“…(These, of course, are the two fundamental, and interconnected, BN modeling challenges in general, not just in the TCGA application.) The latest developments in addressing these two challenges encompass more efficient computational approaches [17,18], and mathematically rigorous and robust methods for handling mixed data, such as mixed local probability models and/or adaptive discretization [18][19][20]. Nevertheless, resolving both difficulties simultaneously in a generalizable toolkit (seamlessly applicable to, for example, across the individual TCGA datasets) remains elusive.…”

Section: Introductionmentioning

confidence: 99%

New analysis framework incorporating mixed mutual information and scalable Bayesian networks for multimodal high dimensional genomic and epigenomic cancer data

Wang

Branciamore

Gogoshin

et al. 2019

Preprint

View full text Add to dashboard Cite

We propose a novel two-stage analysis strategy to discover candidate genes associated with the particular cancer outcomes in large multimodal genomic cancers databases, such as The Cancer Genome Atlas (TCGA). During the first stage, we use mixed mutual information to perform variable selection; during the second stage, we use scalable Bayesian network (BN) modeling to identify candidate genes and their interactions. Two crucial features of the proposed approach are (i) the ability to handle mixed data types (continuous and discrete, genomic, epigenomic, etc.), and (ii) a flexible boundary between the variable selection and network modeling stages ---the boundary that can be adjusted in accordance with the investigators' BN software scalability and hardware implementation. These two aspects result in high generalizability of the proposed analytical framework. We apply the above strategy to three different TCGA datasets (LGG, Brain Lower Grade Glioma; HNSC, Head and Neck Squamous Cell Carcinoma; STES, Stomach and Esophageal Carcinoma), linking multimodal molecular information (SNPs, mRNA expression, DNA methylation) to two clinical outcome variables (tumor status and patient survival). We identify 11 candidate genes, of which 6 have already been directly implicated in the cancer literature. One novel LGG prognostic factor suggested by our analysis, methylation of TMPRSS11F type II transmembrane serine protease, presents intriguing direction for the followup studies.

show abstract

Mixed graphical models for integrative causal analysis with application to chronic lung disease diagnosis and prognosis

Abstract: Supplementary data are available at Bioinformatics online.

Cited by 56 publications

References 40 publications

Bayesian network analysis incorporating genetic anchors complements conventional Mendelian randomization approaches for exploratory analysis of causal relationships in complex data

Bayesian network analysis incorporating genetic anchors complements conventional Mendelian randomization approaches for exploratory analysis of causal relationships in complex data

Synthetic data generation with probabilistic Bayesian Networks

New analysis framework incorporating mixed mutual information and scalable Bayesian networks for multimodal high dimensional genomic and epigenomic cancer data

Contact Info

Product

Resources

About