Combining explainable machine learning, demographic and multi-omic data to inform precision medicine strategies for inflammatory bowel disease

Gardiner, Laura‐Jayne; Carrieri, Anna Paola; Bingham, Karen; Macluskie, Graeme; Bunton, David; McNeil, Marian; Pyzer‐Knapp, Edward O.

doi:10.1371/journal.pone.0263248

Cited by 17 publications

(11 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Prediction of IBD outcomes applying machine learning to these multi-omic data sources represents an exciting direction for future research. 19,49 Conclusions Pediatric CD patients who experience complications show a distinct colonic transcriptome at the time of diagnosis. Machine learning can use this information to predict future outcomes, including strictures, remission, or progression to surgery.…”

Section: Discussionmentioning

confidence: 99%

“…Prediction of IBD outcomes applying machine learning to these multi-omic data sources represents an exciting direction for future research. 19,49…”

Section: Discussionmentioning

confidence: 99%

“…Machine learning techniques, which have the capacity to capture these complex patterns, have been successfully applied to inflammatory bowel disease (IBD)-related topics including identification of risk genes, prediction of outcomes from serum proteins, and prediction of response to medication from multi-omic data. [17][18][19] However, they have not yet been applied specifically to prediction of complications for pediatric CD from gene expression. The goals of our study are twofold: to identify genes which are differentially expressed in CD and complicated CD and to apply machine learning techniques that use those genes to predict risk of complications.…”

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Linking Gene Expression to Clinical Outcomes in Pediatric Crohn’s Disease Using Machine Learning

Chen

Nishiyama²,

Ng³

et al. 2022

Preprint

View full text Add to dashboard Cite

Pediatric Crohn′s disease (CD) is the fastest growing incident age group for the disease, which is characterized by relapsing, remitting inflammation of the gastrointestinal tract and a more frequent occurrence of disease complications such as strictures or fistulas. Previous studies have found intestinal gene expression to be correlated with future disease complications in CD and have developed models to predict risk of these complications. However, they have not focused on colonic gene expression and have not used more complex, machine learning-based modeling techniques. We sought to analyze both ileal and colonic gene expression in a cohort of patients with newly-diagnosed, uncomplicated disease and apply machine learning-based models to predict risk of developing future complications. We included patients from a single institution from 2008-2012 whose biopsies from initial diagnosis were preserved as fresh frozen, paraffin-embedded (FFPE) tissue. Using retrospective chart review, outcomes including strictures (B2), fistulas (B3), progression to surgery, and disease remission were assessed, with a mean follow-up of 6 years. In total, we included 56 patients with CD and 46 controls. Differential analysis revealed a distinct colonic transcriptome for patients who developed strictures, with downregulation of pathways related to inflammation and extra-cellular matrix production. Machine learning-based models were able to incorporate expression of individual genes in combination with clinical characteristics to predict outcomes with high accuracy (area under the receiver operating characteristic curve 0.731 - 0.836). In addition, analysis of the genes most predictive of stricturing revealed several genes (REG1A, MMP3, and DUOX2) with potential biologic importance, which were not evident in single gene differential analysis. Our findings in FFPE tissue support the importance of colonic gene expression and the potential for machine learning-based models in predicting outcomes for pediatric CD.

show abstract

Section: Discussionmentioning

confidence: 99%

“…Prediction of IBD outcomes applying machine learning to these multi-omic data sources represents an exciting direction for future research. 19,49…”

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Linking Gene Expression to Clinical Outcomes in Pediatric Crohn’s Disease Using Machine Learning

Chen

Nishiyama²,

Ng³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“… 71 BIRB796 (doramapimod), another p38 MAPK inhibitor, is used for the treatment of inflammatory bowel diseases. 72 In addition to their therapeutic effects in inflammatory diseases, some p38 MAPK inhibitors, such as neflamapimod, are used in the improvement of cognition and walking disorders. 73 Additionally, p38 MAPK is considered as a key therapeutic target for the treatment of COVID‐19.…”

Section: Kinase Groupsmentioning

confidence: 99%

Role of protein phosphorylation in cell signaling, disease, and the intervention therapy

Pang

Wang

Qin

et al. 2022

MedComm

View full text Add to dashboard Cite

Protein phosphorylation is an important post‐transcriptional modification involving an extremely wide range of intracellular signaling transduction pathways, making it an important therapeutic target for disease intervention. At present, numerous drugs targeting protein phosphorylation have been developed for the treatment of various diseases including malignant tumors, neurological diseases, infectious diseases, and immune diseases. In this review article, we analyzed 303 small‐molecule protein phosphorylation kinase inhibitors (PKIs) registered and participated in clinical research obtained in a database named Protein Kinase Inhibitor Database (PKIDB), including 68 drugs approved by the Food and Drug Administration of the United States. Based on previous classifications of kinases, we divided these human protein phosphorylation kinases into eight groups and nearly 50 families, and delineated their main regulatory pathways, upstream and downstream targets. These groups include: protein kinase A, G, and C (AGC) and receptor guanylate cyclase (RGC) group, calmodulin‐dependent protein kinase (CaMK) group, CMGC [Cyclin‐dependent kinases (CDKs), Mitogen‐activated protein kinases (MAPKs), Glycogen synthase kinases (GSKs), and C dc2‐like kinases (CLKs)] group, sterile (STE)‐MAPKs group, tyrosine kinases (TK) group, tyrosine kinase‐like (TKL) group, atypical group, and other groups. Different groups and families of inhibitors stimulate or inhibit others, forming an intricate molecular signaling regulatory network. This review takes newly developed new PKIs as breakthrough point, aiming to clarify the regulatory network and relationship of each pathway, as well as their roles in disease intervention, and provide a direction for future drug development.

show abstract

Inflammatory bowel disease genomics, transcriptomics, proteomics and metagenomics meet artificial intelligence

Cannarozzi,

Latiano,

Massimino

et al. 2024

UEG Journal

View full text Add to dashboard Cite

Various extrinsic and intrinsic factors such as drug exposures, antibiotic treatments, smoking, lifestyle, genetics, immune responses, and the gut microbiome characterize ulcerative colitis and Crohn's disease, collectively called inflammatory bowel disease (IBD). All these factors contribute to the complexity and heterogeneity of the disease etiology and pathogenesis leading to major challenges for the scientific community in improving management, medical treatments, genetic risk, and exposome impact. Understanding the interaction(s) among these factors and their effects on the immune system in IBD patients has prompted advances in multi‐omics research, the development of new tools as part of system biology, and more recently, artificial intelligence (AI) approaches. These innovative approaches, supported by the availability of big data and large volumes of digital medical datasets, hold promise in better understanding the natural histories, predictors of disease development, severity, complications and treatment outcomes in complex diseases, providing decision support to doctors, and promising to bring us closer to the realization of the “precision medicine” paradigm. This review aims to provide an overview of current IBD omics based on both individual (genomics, transcriptomics, proteomics, metagenomics) and multi‐omics levels, highlighting how AI can facilitate the integration of heterogeneous data to summarize our current understanding of the disease and to identify current gaps in knowledge to inform upcoming research in this field.

show abstract

Combining explainable machine learning, demographic and multi-omic data to inform precision medicine strategies for inflammatory bowel disease

Cited by 17 publications

References 52 publications

Linking Gene Expression to Clinical Outcomes in Pediatric Crohn’s Disease Using Machine Learning

Linking Gene Expression to Clinical Outcomes in Pediatric Crohn’s Disease Using Machine Learning

Role of protein phosphorylation in cell signaling, disease, and the intervention therapy

Inflammatory bowel disease genomics, transcriptomics, proteomics and metagenomics meet artificial intelligence

Contact Info

Product

Resources

About