Machine learning based gut microbiota pattern and response to fiber as a diagnostic tool for chronic inflammatory diseases

Abstract Gut microbiota has been implicated in the pathogenesis of multiple gastrointestinal (GI) and systemic metabolic and inflammatory disorders where disrupted gut microbiota composition and function (dysbiosis) has been found in multiple studies. Thus, human microbiome data holds significant po...

Full description

Saved in:
Bibliographic Details
Main Authors: Miad Boodaghidizaji, Thaisa Jungles, Tingting Chen, Bin Zhang, Tianming Yao, Alan Landay, Ali Keshavarzian, Bruce Hamaker, Arezoo Ardekani
Format: Article
Language:English
Published: BMC 2025-06-01
Series:BMC Microbiology
Subjects:
Online Access:https://doi.org/10.1186/s12866-025-04072-7
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849470226579587072
author Miad Boodaghidizaji
Thaisa Jungles
Tingting Chen
Bin Zhang
Tianming Yao
Alan Landay
Ali Keshavarzian
Bruce Hamaker
Arezoo Ardekani
author_facet Miad Boodaghidizaji
Thaisa Jungles
Tingting Chen
Bin Zhang
Tianming Yao
Alan Landay
Ali Keshavarzian
Bruce Hamaker
Arezoo Ardekani
author_sort Miad Boodaghidizaji
collection DOAJ
description Abstract Gut microbiota has been implicated in the pathogenesis of multiple gastrointestinal (GI) and systemic metabolic and inflammatory disorders where disrupted gut microbiota composition and function (dysbiosis) has been found in multiple studies. Thus, human microbiome data holds significant potential as a source of information for diagnosing and characterizing diseases—such as phenotypes, disease course, and therapeutic response—associated with dysbiotic microbiota communities. However, multiple attempts to leverage gut microbiota taxonomic data for diagnostic and disease characterization have failed due to significant inter-individual variability of microbiota community and overlap of disrupted microbiota communities among multiple diseases. One potential approach is to look at the microbiota community pattern and response to microbiota modifiers like dietary fiber in different disease states. This approach has become feasible with the advent of machine learning, which can uncover hidden patterns in human microbiome data and enable disease prediction. Accordingly, the aim of our study was to test the hypothesis that machine learning algorithms can distinguish stool microbiota patterns—and their responses to fiber—across diseases with previously reported overlapping dysbiotic microbiota profiles. Here, we applied machine learning algorithms to distinguish between Parkinson's disease, Crohn's disease (CD), ulcerative colitis (UC), human immune deficiency virus (HIV), and healthy control (HC) subjects in the presence and absence of fiber treatments. We demonstrated that machine learning algorithms can classify diseases with accuracy as high as 95%. Furthermore, applying machine learning to microbiome data to distinguish UC from CD yielded a prediction accuracy of up to 90%.
format Article
id doaj-art-4fbaa056bf024c5b84f8ddfa9a231aef
institution Kabale University
issn 1471-2180
language English
publishDate 2025-06-01
publisher BMC
record_format Article
series BMC Microbiology
spelling doaj-art-4fbaa056bf024c5b84f8ddfa9a231aef2025-08-20T03:25:12ZengBMCBMC Microbiology1471-21802025-06-0125111110.1186/s12866-025-04072-7Machine learning based gut microbiota pattern and response to fiber as a diagnostic tool for chronic inflammatory diseasesMiad Boodaghidizaji0Thaisa Jungles1Tingting Chen2Bin Zhang3Tianming Yao4Alan Landay5Ali Keshavarzian6Bruce Hamaker7Arezoo Ardekani8School of Mechanical Engineering, Purdue UniversityDepartment of Food Science, Whistler Center for Carbohydrate Research, Purdue UniversityDepartment of Food Science, Whistler Center for Carbohydrate Research, Purdue UniversityDepartment of Food Science, Whistler Center for Carbohydrate Research, Purdue UniversityDepartment of Food Science, Whistler Center for Carbohydrate Research, Purdue UniversityDepartments of Internal Medicine, Anatomy and cell biology, and Molecular Biophysics and Physiology, Center for Integrated Microbiome and Chronobiology Research, Rush University Medical CenterDepartments of Internal Medicine, Anatomy and cell biology, and Molecular Biophysics and Physiology, Center for Integrated Microbiome and Chronobiology Research, Rush University Medical CenterDepartment of Food Science, Whistler Center for Carbohydrate Research, Purdue UniversitySchool of Mechanical Engineering, Purdue UniversityAbstract Gut microbiota has been implicated in the pathogenesis of multiple gastrointestinal (GI) and systemic metabolic and inflammatory disorders where disrupted gut microbiota composition and function (dysbiosis) has been found in multiple studies. Thus, human microbiome data holds significant potential as a source of information for diagnosing and characterizing diseases—such as phenotypes, disease course, and therapeutic response—associated with dysbiotic microbiota communities. However, multiple attempts to leverage gut microbiota taxonomic data for diagnostic and disease characterization have failed due to significant inter-individual variability of microbiota community and overlap of disrupted microbiota communities among multiple diseases. One potential approach is to look at the microbiota community pattern and response to microbiota modifiers like dietary fiber in different disease states. This approach has become feasible with the advent of machine learning, which can uncover hidden patterns in human microbiome data and enable disease prediction. Accordingly, the aim of our study was to test the hypothesis that machine learning algorithms can distinguish stool microbiota patterns—and their responses to fiber—across diseases with previously reported overlapping dysbiotic microbiota profiles. Here, we applied machine learning algorithms to distinguish between Parkinson's disease, Crohn's disease (CD), ulcerative colitis (UC), human immune deficiency virus (HIV), and healthy control (HC) subjects in the presence and absence of fiber treatments. We demonstrated that machine learning algorithms can classify diseases with accuracy as high as 95%. Furthermore, applying machine learning to microbiome data to distinguish UC from CD yielded a prediction accuracy of up to 90%.https://doi.org/10.1186/s12866-025-04072-7Microbiome dataMachine learningUlcerative colitisCrohn's diseaseFiber treatment
spellingShingle Miad Boodaghidizaji
Thaisa Jungles
Tingting Chen
Bin Zhang
Tianming Yao
Alan Landay
Ali Keshavarzian
Bruce Hamaker
Arezoo Ardekani
Machine learning based gut microbiota pattern and response to fiber as a diagnostic tool for chronic inflammatory diseases
BMC Microbiology
Microbiome data
Machine learning
Ulcerative colitis
Crohn's disease
Fiber treatment
title Machine learning based gut microbiota pattern and response to fiber as a diagnostic tool for chronic inflammatory diseases
title_full Machine learning based gut microbiota pattern and response to fiber as a diagnostic tool for chronic inflammatory diseases
title_fullStr Machine learning based gut microbiota pattern and response to fiber as a diagnostic tool for chronic inflammatory diseases
title_full_unstemmed Machine learning based gut microbiota pattern and response to fiber as a diagnostic tool for chronic inflammatory diseases
title_short Machine learning based gut microbiota pattern and response to fiber as a diagnostic tool for chronic inflammatory diseases
title_sort machine learning based gut microbiota pattern and response to fiber as a diagnostic tool for chronic inflammatory diseases
topic Microbiome data
Machine learning
Ulcerative colitis
Crohn's disease
Fiber treatment
url https://doi.org/10.1186/s12866-025-04072-7
work_keys_str_mv AT miadboodaghidizaji machinelearningbasedgutmicrobiotapatternandresponsetofiberasadiagnostictoolforchronicinflammatorydiseases
AT thaisajungles machinelearningbasedgutmicrobiotapatternandresponsetofiberasadiagnostictoolforchronicinflammatorydiseases
AT tingtingchen machinelearningbasedgutmicrobiotapatternandresponsetofiberasadiagnostictoolforchronicinflammatorydiseases
AT binzhang machinelearningbasedgutmicrobiotapatternandresponsetofiberasadiagnostictoolforchronicinflammatorydiseases
AT tianmingyao machinelearningbasedgutmicrobiotapatternandresponsetofiberasadiagnostictoolforchronicinflammatorydiseases
AT alanlanday machinelearningbasedgutmicrobiotapatternandresponsetofiberasadiagnostictoolforchronicinflammatorydiseases
AT alikeshavarzian machinelearningbasedgutmicrobiotapatternandresponsetofiberasadiagnostictoolforchronicinflammatorydiseases
AT brucehamaker machinelearningbasedgutmicrobiotapatternandresponsetofiberasadiagnostictoolforchronicinflammatorydiseases
AT arezooardekani machinelearningbasedgutmicrobiotapatternandresponsetofiberasadiagnostictoolforchronicinflammatorydiseases