Journal Article

A temporal classifier predicts histopathology state and parses acute-chronic phasing in inflammatory bowel disease patients

Peters, Lauren A., Losic, Bojan, Rodrigo Mora, J., Dobrin, Radu, Raymond, Holly A., Friedman, Joshua R., Hagen, Jacob, Curran, Mark, Das, Anuk, Argmann, Carmen, Schadt, Eric E., Chai, Zhi, Restrepo, Paula, Amaro, Mariana P., Xiong, Huabao, Dodatko, Tetyana, Stojmirovic, Aleksandar, Houten, Sander

2023

Overview

Previous studies have conducted time course characterization of murine colitis models through transcriptional profiling of differential expression. We characterize the transcriptional landscape of acute and chronic models of dextran sodium sulfate (DSS) and adoptive transfer (AT) colitis to derive temporal gene expression and splicing signatures in blood and colonic tissue in order to capture dynamics of colitis remission and relapse. We identify sub networks of patient-derived causal networks that are enriched in these temporal signatures to distinguish acute and chronic disease components within the broader molecular landscape of IBD. The interaction between the DSS phenotype and chronological time-point naturally defines parsimonious temporal gene expression and splicing signatures associated with acute and chronic phases disease (as opposed to ordinary time-specific differential expression/splicing). We show these expression and splicing signatures are largely orthogonal, i.e. affect different genetic bodies, and that using machine learning, signatures are predictive of histopathological measures from both blood and intestinal data in murine colitis models as well as an independent cohort of IBD patients. Through access to longitudinal multi-scale profiling from disease tissue in IBD patient cohorts, we can apply this machine learning pipeline to generation of direct patient temporal multimodal regulatory signatures for prediction of histopathological outcomes. Longitudinal changes in gene expression, splicing patterns and adaptive immune repertoire in murine models of colitis are examined and associated with histologic inflammation in mice and in patients with inflammatory bowel disease.

• Publisher: Nature Publishing Group UK,Nature Publishing Group,Nature Portfolio
• Subject: 38/39 / 631/114/1305 / 64 / 64/60 / 692/4020/2199 / Adoptive transfer / Animal models / Animals / Biology / Biomedical and Life Sciences / Chronic illnesses / Colitis / Colitis - chemically induced / Colitis - genetics / Dextran / Dextran Sulfate - toxicity / Gene expression / Inflammatory bowel disease / Inflammatory bowel diseases / Inflammatory Bowel Diseases - genetics / Intestine / Learning algorithms / Life Sciences / Machine learning / Mice / Patients / Phenotype / Phenotypes / Remission / Sodium sulfate / Splicing