Explainable deep learning for stratified medicine in inflammatory bowel disease

Abstract Moving from a one-size-fits-all to an individual approach in precision medicine requires a deeper understanding of disease molecular mechanisms. Especially in heterogeneous complex diseases such as inflammatory bowel disease (IBD), better molecular stratification will help select the correc...

Full description

Saved in:
Bibliographic Details
Main Authors: Nora Verplaetse, Piero Fariselli, Yves Moreau, Daniele Raimondi
Format: Article
Language:English
Published: BMC 2025-07-01
Series:Genome Biology
Subjects:
Online Access:https://doi.org/10.1186/s13059-025-03692-6
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849343890933415936
author Nora Verplaetse
Piero Fariselli
Yves Moreau
Daniele Raimondi
author_facet Nora Verplaetse
Piero Fariselli
Yves Moreau
Daniele Raimondi
author_sort Nora Verplaetse
collection DOAJ
description Abstract Moving from a one-size-fits-all to an individual approach in precision medicine requires a deeper understanding of disease molecular mechanisms. Especially in heterogeneous complex diseases such as inflammatory bowel disease (IBD), better molecular stratification will help select the correct therapy. For this, we build end-to-end biologically sparsified neural network architectures for IBD subtyping based on whole exome sequence representations with gene-level and variant-level resolution. By moving beyond univariate methods, we capitalize on the model’s ability to extract complex molecular patterns to improve prediction. Model interpretation identifies the most predictive pathways, genes, and variants, uncovering important intestinal barrier, immunological, and microbiome factors.
format Article
id doaj-art-8661329b94ea4f66b11f6f70537b44ff
institution Kabale University
issn 1474-760X
language English
publishDate 2025-07-01
publisher BMC
record_format Article
series Genome Biology
spelling doaj-art-8661329b94ea4f66b11f6f70537b44ff2025-08-20T03:42:49ZengBMCGenome Biology1474-760X2025-07-0126113010.1186/s13059-025-03692-6Explainable deep learning for stratified medicine in inflammatory bowel diseaseNora Verplaetse0Piero Fariselli1Yves Moreau2Daniele Raimondi3ESAT-STADIUS, KU LeuvenDepartment of Medical Sciences, University of TorinoESAT-STADIUS, KU LeuvenESAT-STADIUS, KU LeuvenAbstract Moving from a one-size-fits-all to an individual approach in precision medicine requires a deeper understanding of disease molecular mechanisms. Especially in heterogeneous complex diseases such as inflammatory bowel disease (IBD), better molecular stratification will help select the correct therapy. For this, we build end-to-end biologically sparsified neural network architectures for IBD subtyping based on whole exome sequence representations with gene-level and variant-level resolution. By moving beyond univariate methods, we capitalize on the model’s ability to extract complex molecular patterns to improve prediction. Model interpretation identifies the most predictive pathways, genes, and variants, uncovering important intestinal barrier, immunological, and microbiome factors.https://doi.org/10.1186/s13059-025-03692-6Genome interpretationNeural networksMachine learning
spellingShingle Nora Verplaetse
Piero Fariselli
Yves Moreau
Daniele Raimondi
Explainable deep learning for stratified medicine in inflammatory bowel disease
Genome Biology
Genome interpretation
Neural networks
Machine learning
title Explainable deep learning for stratified medicine in inflammatory bowel disease
title_full Explainable deep learning for stratified medicine in inflammatory bowel disease
title_fullStr Explainable deep learning for stratified medicine in inflammatory bowel disease
title_full_unstemmed Explainable deep learning for stratified medicine in inflammatory bowel disease
title_short Explainable deep learning for stratified medicine in inflammatory bowel disease
title_sort explainable deep learning for stratified medicine in inflammatory bowel disease
topic Genome interpretation
Neural networks
Machine learning
url https://doi.org/10.1186/s13059-025-03692-6
work_keys_str_mv AT noraverplaetse explainabledeeplearningforstratifiedmedicineininflammatoryboweldisease
AT pierofariselli explainabledeeplearningforstratifiedmedicineininflammatoryboweldisease
AT yvesmoreau explainabledeeplearningforstratifiedmedicineininflammatoryboweldisease
AT danieleraimondi explainabledeeplearningforstratifiedmedicineininflammatoryboweldisease