Machine learning and SHAP-based identification of RNASE1 linking environmental endocrine-disrupting chemicals exposure to atherosclerosis

Environmental endocrine-disrupting chemicals (EDCs) are ubiquitous pollutants implicated in cardiometabolic disorders, yet their mechanistic contribution to atherosclerosis (AS) remains elusive. Seven prevalent EDCs - bisphenol A, dibutyl phthalate, di(2-ethylhexyl) phthalate, dioxins, polychlorinated biphenyls (PCBs), perfluorooctanoic acid and PFOS - were selected. Network toxicology integrated ChEMBL, STITCH and SwissTargetPrediction to compile human EDCs targets. Differentially expressed genes were derived from 3 Gene Expression Omnibus AS datasets after batch correction. Key genes were filtered by least absolute shrinkage and selection operator regression and support vector machine-recursive feature elimination (SVM-RFE) and intersected with EDCs targets. Diagnostic performance was evaluated by receiver operating characteristic analysis; functional relevance was assessed by KEGG-GSEA, and feature importance was quantified with SHapley Additive exPlanations (SHAP). A total of 1294 nonredundant EDCs target genes were enriched in lipid and AS and endocrine resistance pathways (FDR < 0.05). From 789 AS differentially expressed genes, least absolute shrinkage and selection operator and SVM-RFE converged on 11 robust candidates. Intersection with EDCs targets pinpointed NTRK3 and RNASE1. Receiver operating characteristic analysis yielded AUCs of 0.777 and 0.859, respectively, with a combined AUC of 0.871. GSEA indicated enrichment of cytokine/chemokine and NOD-like receptor signaling for both genes. SHAP scores highlighted RNASE1 as the dominant predictor (mean SHAP = 0.215), whereas NTRK3 contributed modestly (mean SHAP = 0.005). RNASE1 and NTRK3 were identified as key molecular links between endocrine-disrupting chemical exposure and AS. Among them, RNASE1 was revealed as a novel and dominant predictor, highlighting its unique mechanistic role and potential clinical utility in environmental cardiovascular risk stratification and targeted prevention.

Authors

Institutions

Publication Details

Journal
Medicine
Published
2025-09-12
DOI
https://doi.org/10.1097/md.0000000000044567
Citations
1
Primary Topic
Effects and risks of endocrine disrupting chemicals
Type
article
Field-Weighted Citation Impact
1.20
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Machine learning and SHAP-based identification of RNASE1 linking environmental endocrine-disrupting chemicals exposure to atherosclerosis

Qiang Xiao, Xin Wen Guo, Weinan Wang, Xin Zhou et al.
1 citations
Medicine
Effects and risks of endocrine disrupting chemicals
1.20
article

Machine learning and SHAP-based identification of RNASE1 linking environmental endocrine-disrupting chemicals exposure to atherosclerosis

Qiang Xiao, Xin Wen Guo, Weinan Wang, Xin Zhou, Mou‐Tuan Huang
article en
1 citations

Abstract

Environmental endocrine-disrupting chemicals (EDCs) are ubiquitous pollutants implicated in cardiometabolic disorders, yet their mechanistic contribution to atherosclerosis (AS) remains elusive. Seven prevalent EDCs - bisphenol A, dibutyl phthalate, di(2-ethylhexyl) phthalate, dioxins, polychlorinated biphenyls (PCBs), perfluorooctanoic acid and PFOS - were selected. Network toxicology integrated ChEMBL, STITCH and SwissTargetPrediction to compile human EDCs targets. Differentially expressed genes were derived from 3 Gene Expression Omnibus AS datasets after batch correction. Key genes were filtered by least absolute shrinkage and selection operator regression and support vector machine-recursive feature elimination (SVM-RFE) and intersected with EDCs targets. Diagnostic performance was evaluated by receiver operating characteristic analysis; functional relevance was assessed by KEGG-GSEA, and feature importance was quantified with SHapley Additive exPlanations (SHAP). A total of 1294 nonredundant EDCs target genes were enriched in lipid and AS and endocrine resistance pathways (FDR < 0.05). From 789 AS differentially expressed genes, least absolute shrinkage and selection operator and SVM-RFE converged on 11 robust candidates. Intersection with EDCs targets pinpointed NTRK3 and RNASE1. Receiver operating characteristic analysis yielded AUCs of 0.777 and 0.859, respectively, with a combined AUC of 0.871. GSEA indicated enrichment of cytokine/chemokine and NOD-like receptor signaling for both genes. SHAP scores highlighted RNASE1 as the dominant predictor (mean SHAP = 0.215), whereas NTRK3 contributed modestly (mean SHAP = 0.005). RNASE1 and NTRK3 were identified as key molecular links between endocrine-disrupting chemical exposure and AS. Among them, RNASE1 was revealed as a novel and dominant predictor, highlighting its unique mechanistic role and potential clinical utility in environmental cardiovascular risk stratification and targeted prevention.

MedicineVol. 104(37)
Ganzhou People's Hospital (CN), Gannan Medical University (CN), Jiangxi Pingxiang People's Hospital (CN), First Affiliated Hospital of Gannan Medical University (CN)
Openalex Percentile: Top 18%
Effects and risks of endocrine disrupting chemicals
1.20
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.