Nat Commun. 2021 Jul 12;12(1):4246. doi: 10.1038/s41467-021-24470-5.

Debashis Sahoo ^# 1 2 3, Lee Swanson ^# 4, Ibrahim M Sayed ^# 5 6, Gajanan D Katkar ⁴, Stella-Rita Ibeawuchi ⁶, Yash Mittal ⁷, Rama F Pranadinata ⁴, Courtney Tindle ⁴, Mackenzie Fuller ⁴, Dominik L Stec ⁴, John T Chang ⁷, William J Sandborn ⁷, Soumita Das ⁸, Pradipta Ghosh ^{9 10 11 12}

Author information

• ¹Department of Pediatrics, University of California San Diego, San Diego, CA, USA. dsahoo@ucsd.edu.
• ²Department of Computer Science and Engineering, Jacob's School of Engineering, University of California San Diego, San Diego, CA, USA. dsahoo@ucsd.edu.
• ³Rebecca and John Moore Comprehensive Cancer Center, University of California San Diego, San Diego, CA, USA. dsahoo@ucsd.edu.
• ⁴Department of Cellular and Molecular Medicine, University of California San Diego, San Diego, CA, USA.
• ⁵Department of Medical Microbiology and Immunology, Faculty of Medicine, Assiut University, Assiu, Egypt.
• ⁶Department of Pathology, University of California San Diego, San Diego, CA, USA.
• ⁷Department of Medicine, University of California San Diego, San Diego, CA, USA.
• ⁸Department of Pathology, University of California San Diego, San Diego, CA, USA. sodas@ucsd.edu.
• ⁹Rebecca and John Moore Comprehensive Cancer Center, University of California San Diego, San Diego, CA, USA. prghosh@ucsd.edu.
• ¹⁰Department of Cellular and Molecular Medicine, University of California San Diego, San Diego, CA, USA. prghosh@ucsd.edu.
• ¹¹Department of Medicine, University of California San Diego, San Diego, CA, USA. prghosh@ucsd.edu.
• ¹²Veterans Affairs Medical Center, La Jolla, CA, USA. prghosh@ucsd.edu.

^#Contributed equally.

Abstract

Modeling human diseases as networks simplify complex multi-cellular processes, helps understand patterns in noisy data that humans cannot find, and thereby improves precision in prediction. Using Inflammatory Bowel Disease (IBD) as an example, here we outline an unbiased AI-assisted approach for target identification and validation. A network was built in which clusters of genes are connected by directed edges that highlight asymmetric Boolean relationships. Using machine-learning, a path of continuum states was pinpointed, which most effectively predicted disease outcome. This path was enriched in gene-clusters that maintain the integrity of the gut epithelial barrier. We exploit this insight to prioritize one target, choose appropriate pre-clinical murine models for target validation and design patient-derived organoid models. Potential for treatment efficacy is confirmed in patient-derived organoids using multivariate analyses. This AI-assisted approach identifies a first-in-class gut barrier-protective agent in IBD and predicted Phase-III success of candidate agents.