Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
September 15, 2023Biomolecules and BiomedicineOpen Access

A novel machine learning-derived four-gene signature predicts STEMI and post-STEMI heart failure

View Full Paper
Ask AI
Bookmark
Share

Key result

A machine learning-derived four-gene panel (HLA-J, CFP, STX11, and NFYC) accurately discriminated STEMI and post-STEMI heart failure with an area under the curve ranging from 0.86 to 0.91.

Why the study?

High mortality and morbidity from STEMI and post-STEMI HF necessitate accurate CAD risk stratification, requiring a specific and convenient prediction model.

Does a four-gene signature panel predict the risk of STEMI and post-STEMI heart failure in patients with coronary artery disease?

Population

1,956 monocyte expression profiles and clinical data integrated from multiple sources

Comparison

Machine learning models based on DT, SVM, and RF algorithms

Design

Machine learning model development and validation study

Authors

JYJialu YaoYZYujia ZhouZYZhichao Yao

Discussion

Loading...

Member takes

Overview

May aid monocyte gene-based STEMI risk stratification; hypothesis-generating and requires prospective validation before any clinical use.

Key Points

  • To develop a monocyte-based gene expression assay utilizing machine learning to predict ST-elevation myocardial infarction (STEMI) and subsequent post-STEMI heart failure.
  • Integrated 1,956 monocyte expression profiles alongside corresponding clinical data from multiple sources.
  • Identified candidate genes using weighted gene co-expression network analysis (WGCNA) and differential expression analysis.
  • Trained and evaluated machine learning classifiers using decision tree, support vector machine, and random forest algorithms.
  • Decision tree modeling identified a four-gene panel (HLA-J, CFP, STX11, and NFYC) that outperformed support vector machine and random forest classifiers.
  • The four-gene panel discriminated STEMI and post-STEMI heart failure with an area under the curve (AUC) of 0.86 or higher.
  • Gene set enrichment analysis demonstrated significant, concordant differences in cardiac pathogenesis pathways between groups categorized by panel expression levels.

Study Design

Type

Observational (n=1,956)

Structured PICO

Does a four-gene signature panel predict the risk of STEMI and post-STEMI heart failure in patients with coronary artery disease?

P
Population
1,956 monocyte expression profiles from multiple retrospective cohorts were analyzed to develop and validate a machine learning model for predicting STEMI and post-STEMI heart failure.
E
Exposure
A machine learning-derived four-gene signature panel (HLA-J, CFP, STX11, and NFYC) based on a decision tree algorithm.
O
Outcome
Discrimination of STEMI and post-STEMI heart failure risk.surrogate

Main Result

Effect estimate: AUC 0.86-0.91

A novel machine learning-derived four-gene signature (HLA-J, CFP, STX11, and NFYC) demonstrates high accuracy in predicting STEMI and post-STEMI heart failure, offering a potential tool for risk stratification.

Limitations

  • Lack of fully independent external validation (only split-sample internal validation was performed)
  • Small sample size for heart failure risk prediction
  • Uniform ethnicity composition (mainly Caucasian) from Europe and US datasets
  • Retrospective nature of the datasets used
  • Internal validation only (split-sample validation), lacking fully independent external validation
  • Small sample size for model development and validation
  • Small subject number for heart failure risk prediction

Cite This Study

Yao et al. (2023) conducted an observational in ST-elevation myocardial infarction (STEMI) and post-STEMI heart failure (n=1,956). Four-gene signature (HLA-J, CFP, STX11, NFYC) vs. Stable CAD or healthy controls was evaluated on Discrimination of STEMI and post-STEMI heart failure (AUC 0.86-0.91). A machine learning-derived four-gene panel (HLA-J, CFP, STX11, and NFYC) accurately discriminated STEMI and post-STEMI heart failure with an area under the curve ranging from 0.86 to 0.91.

synapsesocial.com/papers/6a75be92f1e99ff85ab2107chttps://doi.org/10.17305/bb.2023.9629
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Integrated Gene Expression Profiling Analysis Reveals Potential Molecular Mechanisms and Candidate Biomarkers for Early Risk Stratification and Prediction of STEMI and Post-STEMI Heart Failure Patients2020 · 1 citations
  2. 2Integrated Gene Expression Profiling Analysis Reveals Potential Molecular Mechanisms and Candidate Biomarkers for Early Risk Stratification and Prediction of STEMI and Post-STEMI Heart Failure Patients2021 · 16 citations
  3. 3Identification of monocyte-associated genes as predictive biomarkers of heart failure after acute myocardial infarction2021 · 29 citations
  4. 4Exploring an immune cells-related molecule in STEMI by bioinformatics analysis2023 · 9 citations
  5. 5Identification of signature genes and subtypes for heart failure diagnosis based on machine learning2025 · 5 citations