Abstract
Research of inflammatory bowel disease (IBD) has identified numerous molecular players involved in the disease development. Even so, the understanding of IBD is incomplete, while disease treatment is still far from the precision medicine. Reliable diagnostic and prognostic biomarkers in IBD are limited which may reduce efficient therapeutic outcomes. High-throughput technologies and artificial intelligence emerged as powerful tools in search of unrevealed molecular patterns that could give important insights into IBD pathogenesis and help to address unmet clinical needs. Machine learning, a subtype of artificial intelligence, uses complex mathematical algorithms to learn from existing data in order to predict future outcomes. The scientific community has been increasingly employing machine learning for the prediction of IBD outcomes from comprehensive patient data-clinical records, genomic, transcriptomic, proteomic, metagenomic, and other IBD relevant omics data. This review aims to present fundamental principles behind machine learning modeling and its current application in IBD research with the focus on studies that explored genomic and transcriptomic data. We described different strategies used for dealing with omics data and outlined the best-performing methods. Before being translated into clinical settings, the developed machine learning models should be tested in independent prospective studies as well as randomized controlled trials.
Highlights
Inflammatory bowel disease (IBD) is a complex disease, characterized as chronic, relapsing and remitting intestinal inflammation, with substantial heterogeneity among clinical phenotypes with regards to the age at diagnosis, severity of symptoms, response to therapy and long-term clinical outcomes [1,2,3]
Example data contains labeled instances, which means that both input and output, which is the phenotype of interest, are known
genomeorwide association studies (GWAS)—genome-wide association study, IBD—inflammatory bowel disease, CD—Crohn’s disease, UC—ulcerative colitis, support vector machines (SVM)—support vector machine, AUC—area under the receiver operating curve, AUPRC—area under the precision-recall curve, IIBDGC—The International Inflammatory Bowel Disease Genetics Consortium, *—correlation of cell type frequencies between hieratical clustering analysis applied to RNA profile of a cell and cytometry results referring to that cell
Summary
Inflammatory bowel disease (IBD) is a complex disease, characterized as chronic, relapsing and remitting intestinal inflammation, with substantial heterogeneity among clinical phenotypes with regards to the age at diagnosis, severity of symptoms, response to therapy and long-term clinical outcomes [1,2,3]. Regarding molecular classification of disease subtypes, it has been shown that most of detected IBD loci confer risk to both CD and UC, typically with distinct effect sizes in each disorder; whereas the minor number of loci is unique to each subtype [15,30] In addition to the latter, there are examples, such as variants at the NOD2 and protein tyrosine phosphatase nonreceptor type 22 (PTPN22) loci, which have been found to be risk factors for CD, while for UC they have been shown to be variants with a protective effect [30]. Important additional evidence is that differential microRNA (miRNA) expression was detected between these entities [31] In this sense, it has been suggested recently that ileal and colonic CD could potentially be regarded as separate diseases and that consideration should be given to a new classification for CD, which splits it into ileum dominant (isolated ileal and ileocolonic).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
