Population Cohort-Validated PM<sub>2.5</sub>-Induced Gene Signatures: A Machine Learning Approach to Individual Exposure Prediction

Transcriptomic profiling has shown that exposure to PM<sub>2.5</sub>, a common air pollutant, can modulate gene expression, which has been linked to negative health effects and diseases. However, there are few population-based cohort studies on the association between PM<sub>2.5<...

Full description

Saved in:
Bibliographic Details
Main Authors: Yu-Chung Wei, Wen-Chi Cheng, Pinpin Lin, Zhi-Yao Zhang, Chi-Hsien Chen, Chih-Da Wu, Yue Leon Guo, Hung-Jung Wang
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Toxics
Subjects:
Online Access:https://www.mdpi.com/2305-6304/13/7/562
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849251805664378880
author Yu-Chung Wei
Wen-Chi Cheng
Pinpin Lin
Zhi-Yao Zhang
Chi-Hsien Chen
Chih-Da Wu
Yue Leon Guo
Hung-Jung Wang
author_facet Yu-Chung Wei
Wen-Chi Cheng
Pinpin Lin
Zhi-Yao Zhang
Chi-Hsien Chen
Chih-Da Wu
Yue Leon Guo
Hung-Jung Wang
author_sort Yu-Chung Wei
collection DOAJ
description Transcriptomic profiling has shown that exposure to PM<sub>2.5</sub>, a common air pollutant, can modulate gene expression, which has been linked to negative health effects and diseases. However, there are few population-based cohort studies on the association between PM<sub>2.5</sub> exposure and specific gene set expression. In this study, we used an unbiased transcriptomic profiling approach to examine gene expression in a mouse model exposed to PM<sub>2.5</sub> and to identify PM<sub>2.5</sub>-responsive genes. The gene expressions were further validated in both the human cell lines and a population-based cohort study. Two cohorts of healthy older adults (aged ≥ 65 years) were recruited from regions characterized by differing levels of PM<sub>2.5</sub>. Logistic regression and decision tree algorithms were then utilized to construct predictive models for PM<sub>2.5</sub> exposure based on these gene expression profiles. Our results indicated that the expression of five genes (<i>FAM102B</i>, <i>PPP2R1B</i>, <i>OXR1</i>, <i>ITGAM</i>, and <i>PRP38B)</i> increased with PM<sub>2.5</sub> exposure in both cell-based assay and population-based cohort studies. Furthermore, the predictive models demonstrated high accuracy in classifying high-and-low PM<sub>2.5</sub> exposure, potentially supporting the integration of gene biomarkers into public health practices.
format Article
id doaj-art-ca22c1b22bff4670a4abfea21df7f491
institution Kabale University
issn 2305-6304
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Toxics
spelling doaj-art-ca22c1b22bff4670a4abfea21df7f4912025-08-20T03:56:49ZengMDPI AGToxics2305-63042025-06-0113756210.3390/toxics13070562Population Cohort-Validated PM<sub>2.5</sub>-Induced Gene Signatures: A Machine Learning Approach to Individual Exposure PredictionYu-Chung Wei0Wen-Chi Cheng1Pinpin Lin2Zhi-Yao Zhang3Chi-Hsien Chen4Chih-Da Wu5Yue Leon Guo6Hung-Jung Wang7Graduate Institute of Statistics and Information Science, National Changhua University of Education, Changhua City 500207, TaiwanInstitute of Medical Sciences, Tzu Chi University, Hualien City 970374, TaiwanNational Institute of Environmental Health Sciences, National Health Research Institutes, Zhunan 35041, TaiwanInstitute of Biomedical Sciences and Engineering, Tzu Chi University, Hualien City 970374, TaiwanDepartment of Environmental and Occupational Medicine, National Taiwan University College of Medicine and National Taiwan University Hospital, Taipei 100225, TaiwanNational Institute of Environmental Health Sciences, National Health Research Institutes, Zhunan 35041, TaiwanNational Institute of Environmental Health Sciences, National Health Research Institutes, Zhunan 35041, TaiwanInstitute of Medical Sciences, Tzu Chi University, Hualien City 970374, TaiwanTranscriptomic profiling has shown that exposure to PM<sub>2.5</sub>, a common air pollutant, can modulate gene expression, which has been linked to negative health effects and diseases. However, there are few population-based cohort studies on the association between PM<sub>2.5</sub> exposure and specific gene set expression. In this study, we used an unbiased transcriptomic profiling approach to examine gene expression in a mouse model exposed to PM<sub>2.5</sub> and to identify PM<sub>2.5</sub>-responsive genes. The gene expressions were further validated in both the human cell lines and a population-based cohort study. Two cohorts of healthy older adults (aged ≥ 65 years) were recruited from regions characterized by differing levels of PM<sub>2.5</sub>. Logistic regression and decision tree algorithms were then utilized to construct predictive models for PM<sub>2.5</sub> exposure based on these gene expression profiles. Our results indicated that the expression of five genes (<i>FAM102B</i>, <i>PPP2R1B</i>, <i>OXR1</i>, <i>ITGAM</i>, and <i>PRP38B)</i> increased with PM<sub>2.5</sub> exposure in both cell-based assay and population-based cohort studies. Furthermore, the predictive models demonstrated high accuracy in classifying high-and-low PM<sub>2.5</sub> exposure, potentially supporting the integration of gene biomarkers into public health practices.https://www.mdpi.com/2305-6304/13/7/562PM<sub>2.5</sub>transcriptomic profilingbiomarkermachine learningpredictive model
spellingShingle Yu-Chung Wei
Wen-Chi Cheng
Pinpin Lin
Zhi-Yao Zhang
Chi-Hsien Chen
Chih-Da Wu
Yue Leon Guo
Hung-Jung Wang
Population Cohort-Validated PM<sub>2.5</sub>-Induced Gene Signatures: A Machine Learning Approach to Individual Exposure Prediction
Toxics
PM<sub>2.5</sub>
transcriptomic profiling
biomarker
machine learning
predictive model
title Population Cohort-Validated PM<sub>2.5</sub>-Induced Gene Signatures: A Machine Learning Approach to Individual Exposure Prediction
title_full Population Cohort-Validated PM<sub>2.5</sub>-Induced Gene Signatures: A Machine Learning Approach to Individual Exposure Prediction
title_fullStr Population Cohort-Validated PM<sub>2.5</sub>-Induced Gene Signatures: A Machine Learning Approach to Individual Exposure Prediction
title_full_unstemmed Population Cohort-Validated PM<sub>2.5</sub>-Induced Gene Signatures: A Machine Learning Approach to Individual Exposure Prediction
title_short Population Cohort-Validated PM<sub>2.5</sub>-Induced Gene Signatures: A Machine Learning Approach to Individual Exposure Prediction
title_sort population cohort validated pm sub 2 5 sub induced gene signatures a machine learning approach to individual exposure prediction
topic PM<sub>2.5</sub>
transcriptomic profiling
biomarker
machine learning
predictive model
url https://www.mdpi.com/2305-6304/13/7/562
work_keys_str_mv AT yuchungwei populationcohortvalidatedpmsub25subinducedgenesignaturesamachinelearningapproachtoindividualexposureprediction
AT wenchicheng populationcohortvalidatedpmsub25subinducedgenesignaturesamachinelearningapproachtoindividualexposureprediction
AT pinpinlin populationcohortvalidatedpmsub25subinducedgenesignaturesamachinelearningapproachtoindividualexposureprediction
AT zhiyaozhang populationcohortvalidatedpmsub25subinducedgenesignaturesamachinelearningapproachtoindividualexposureprediction
AT chihsienchen populationcohortvalidatedpmsub25subinducedgenesignaturesamachinelearningapproachtoindividualexposureprediction
AT chihdawu populationcohortvalidatedpmsub25subinducedgenesignaturesamachinelearningapproachtoindividualexposureprediction
AT yueleonguo populationcohortvalidatedpmsub25subinducedgenesignaturesamachinelearningapproachtoindividualexposureprediction
AT hungjungwang populationcohortvalidatedpmsub25subinducedgenesignaturesamachinelearningapproachtoindividualexposureprediction