DAGBagM: learning directed acyclic graphs of mixed variables with an application to identify protein biomarkers for treatment response in ovarian cancer

• Published in: BMC bioinformatics Vol. 23; no. 1; pp. 321 - 19
• Main Authors: Chowdhury, Shrabanti, Wang, Ru, Yu, Qing, Huntoon, Catherine J., Karnitz, Larry M., Kaufmann, Scott H., Gygi, Steven P., Birrer, Michael J., Paulovich, Amanda G., Peng, Jie, Wang, Pei
• Format: Journal Article
• Language: English
• Published: London BioMed Central 05.08.2022; BioMed Central Ltd; BMC
• Subjects: Algorithms; Bioinformatics; Biomarkers; Biomedical and Life Sciences; Bootstrap aggregation; Cancer; Cancer therapies; Care and treatment; Causality; Child; Computational Biology/Bioinformatics; Computer Appl. in Life Sciences; Computer Simulation; Confounding Factors, Epidemiologic; Continuity (mathematics); Development and progression; Diagnosis; Experiments; Female; Graph theory; Hill climbing; Humans; Identification and classification; Learning; Life Sciences; Methods; Microarrays; Nodes; Oncology, Experimental; Ovarian cancer; Ovarian Neoplasms - drug therapy; Ovarian Neoplasms - genetics; Performance evaluation; Proteins; Proteomics; Random variables; Sensitive and resistant/refractory; Simulation; Tumor markers; United States; Hill climbing; Bootstrap aggregation; Sensitive and resistant/refractory; Proteomics
• ISSN: 1471-2105; 1471-2105
• DOI: 10.1186/s12859-022-04864-y

Background Applying directed acyclic graph (DAG) models to proteogenomic data has been shown effective for detecting causal biomarkers of complex diseases. However, there remain unsolved challenges in DAG learning to jointly model binary clinical outcome variables and continuous biomarker measurements. Results In this paper, we propose a new tool, DAGBagM, to learn DAGs with both continuous and binary nodes. By using appropriate models, DAGBagM allows for either continuous or binary nodes to be parent or child nodes. It employs a bootstrap aggregating strategy to reduce false positives in edge inference. At the same time, the aggregation procedure provides a flexible framework to robustly incorporate prior information on edges. Conclusions Through extensive simulation experiments, we demonstrate that DAGBagM has superior performance compared to alternative strategies for modeling mixed types of nodes. In addition, DAGBagM is computationally more efficient than two competing methods. When applying DAGBagM to proteogenomic datasets from ovarian cancer studies, we identify potential protein biomarkers for platinum refractory/resistant response in ovarian cancer. DAGBagM is made available as a github repository at https://github.com/jie108/dagbagM .