Machine learning-based integration develops an immune-derived signature for diagnosing high-altitude pulmonary hypertension - Original Research

Q: Why is early diagnosis of high-altitude pulmonary hypertension difficult in high-altitude regions?

Gold-standard diagnosis relies on right heart catheterization and specialized infrastructure, and initial clinical symptoms are often nonspecific, causing delays and reduced therapeutic effectiveness.

Q: What data types were integrated to build the diagnostic model?

The study used single-cell RNA sequencing, bulk RNA sequencing, and proteomic profiling, then combined them through an integrative multi-omics workflow.

Q: How was the immune-derived signature validated and what model achieved the best performance?

HAPH-associated biomarkers were refined via ensemble machine learning and validated by quantitative PCR; a six-gene random forest model achieved high diagnostic accuracy with an AUC of 0.995 in training and 0.773 in external validation.

Aug 4, 2026

Upload by Clementine

High-altitude pulmonary hypertension (HAPH) requires early, accurate diagnosis, yet gold-standard procedures remain difficult to deploy in high-altitude settings. A retrospective study integrates single-cell RNA sequencing, bulk RNA sequencing, and proteomic profiling to map immune microenvironment remodeling and train a machine learning diagnostic model. HAPH-associated signatures are validated with quantitative PCR, and a six-gene random forest model shows strong performance across training and external validation cohorts.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Introduction

Background and diagnostic challenges

Current diagnostic standards and unmet needs

Methods

Integrative multi-omics cohort and retrospective design

Machine learning model construction and validation

Results

Immune lineage roles and gene discovery

Differential expression and biomarker screening

Model performance and quantitative PCR validation

Conclusion

Blood-derived immune signature for early detection

OPEN ACCESS EDITED BY Dawei Yang, Fudan University, China REVIEWED BY Mayur Doke, University of Miami, United States Yunhuan Liu, Tongji University, China *CORRESPONDENCE Wu Li [goodli002@163.com](goodli002@163.com)[ ](goodli002@163.com)Dongfeng Yin [ydf1112@163.com](ydf1112@163.com) †These authors have contributed equally to this work RECEIVED 15 April 2025 ACCEPTED 15 August 2025 PUBLISHED 02 September 2025 CITATION Yang D, Li Q, Yang F, Wang R, Jiang P, Wu J, Yang X, Huang Y, Liu Y, Wang S, Gou J, Sun Z, Ma J, Qin Y, Li W and Yin D (2025) Machine learning-based integration develops an immune-derived signature for diagnosing high-altitude pulmonary hypertension. Front. Med. 12:1603140 . doi: 10.3389/fmed.2025.1603140 COPYRIGHT © 2025 Yang, Li, Yang, Wang, Jiang, Wu, Yang, Huang, Liu, Wang, Gou, Sun, Ma, Qin, Li and Yin. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. TYPE Original Research PUBLISHED 02 September 2025 DOI 10.3389/fmed.2025.1603140 Machine learning-based integration develops an immune-derived signature for diagnosing high-altitude pulmonary hypertension Dan Yang 1,2†, Qian Li 1,2†, Feng Yang 1†, Rui Wang 1, Peng Jiang 1, Jialin Wu 1, Xi Yang 1, Yixuan Huang 2, Yuqiang Liu3, Shishang Wang 1, Junqiang Gou 1, Zhangfeng Sun 2, Junjie Ma 1, Yanhui Qin 2, Wu Li 1* and Dongfeng Yin 1,2,4* 1General Hospital of Xinjiang Military Command, Urumqi, China, 2Xinjiang Medical University, Urumqi, China, 3 No.951 Hospital of PLA, Korla, China, 4Shihezi University, Shihezi, China Background: High-altitude pulmonary hypertension (HAPH) is a common disease in high-altitude regions where implementation of gold-standard diagnostic methods remains logistically challenging. Methods: In the retrospective analysis, we employed an integrative multi-omics approach combining single-cell RNA sequencing (scRNA-seq, n = 10), bulk RNA sequencing (RNA-seq, n = 126), and proteomic profiling (n = 42) to characterize immune microenvironment remodeling in HAPH. Subsequently, we established a machine learning-based diagnostic model. The HAPH-associated signatures were finally validated by Quantitative PCR. Results: Through scRNA-seq analysis utilizing Ro/e and contribution scoring analysis, we first demonstrated the pivotal role of myeloid lineages in HAPHpathogenesis. Pseudotime trajectory analysis of the myeloid subsets further revealed 2,615 differentially expressed genes (DEGs) associated with HAPH progression. We also identified 144 and 77 DEGs from bulk RNA-seq and proteomic data between HAPH and control groups, respectively. Finally, 22 candidate biomarkers were screened by muti-omics analysis. These genes were further refined through ensemble machine learning algorithms. Evaluation of 113 algorithm combinations revealed that a six-gene random forest (RF) model (HEMGN, HBG2, MYL9, ANK1, UBE2O, RBPMS2) achieved optimal diagnostic accuracy, with an area under the curve (AUC) of 0.995 in the training cohort (n = 55) and 0.773 in external validation cohorts (n = 71) . Quantitative PCR validated significant overexpression of these biomarkers in HAPH compared to controls (p < 0.05) . Conclusion: Our findings propose the minimally invasive blood-derived immune signature for HAPH diagnosis, providing a practical framework for early detection in resource-constrained high-altitude populations. KEYWORDS high-altitude pulmonary hypertension, single-cell RNA sequencing, multi-omics integration, machine learning, non-invasive diagnosis Frontiers in Medicine 01 [frontiersin.org](frontiersin.or

FAQs

Why is early diagnosis of high-altitude pulmonary hypertension difficult in high-altitude regions?

Gold-standard diagnosis relies on right heart catheterization and specialized infrastructure, and initial clinical symptoms are often nonspecific, causing delays and reduced therapeutic effectiveness.

What data types were integrated to build the diagnostic model?

The study used single-cell RNA sequencing, bulk RNA sequencing, and proteomic profiling, then combined them through an integrative multi-omics workflow.

How was the immune-derived signature validated and what model achieved the best performance?

HAPH-associated biomarkers were refined via ensemble machine learning and validated by quantitative PCR; a six-gene random forest model achieved high diagnostic accuracy with an AUC of 0.995 in training and 0.773 in external validation.

Aug 4, 2026

Upload by Clementine