What is the clinical evidence from this study?

Study design: Observational. Population: ST-segment elevation myocardial infarction (STEMI) and post-STEMI heart failure. Intervention: 5-gene machine learning prediction model (SLC2A3, CLEC4D, GPR97, PLAUR, BST1) vs. Healthy controls. Primary outcome: Prediction of STEMI (combination set) (AUC 0.9441).

December 10, 2020Open Access

Integrated Gene Expression Profiling Analysis Reveals Potential Molecular Mechanisms and Candidate Biomarkers for Early Risk Stratification and Prediction of STEMI and Post-STEMI Heart Failure Patients

View Full Paper

Ask AI

Bookmark

Key result

Five-gene machine learning model predicts STEMI with ~0.94 AUC and identifies post-STEMI HF biomarkers.

AUC 0.9441

Why the study?

To explore the molecular mechanisms and identify candidate whole-blood predictive and prognostic biomarkers in patients with STEMI and post-STEMI heart failure.

Do specific peripheral blood gene expression profiles predict the occurrence of STEMI and subsequent post-STEMI heart failure?

Population

Whole blood gene expression datasets GSE60993, GSE61144, GSE66360, and GSE59867 from NCBI-GEO

Design

Integrated bioinformatics and machine learning analysis

Authors

JXJing Xu YYYuejing Yang

Discussion

Member takes

Overview

May support blood-based biomarkers for STEMI risk; leaves open clinical adoption pending prospective validation.

Study Design

Type

Observational

Structured PICO

Do specific peripheral blood gene expression profiles predict the occurrence of STEMI and subsequent post-STEMI heart failure?

Population

Microarray data profiles of STEMI patients, post-STEMI heart failure (HF) patients, and healthy controls from four GEO datasets (GSE60993, GSE61144, GSE66360, GSE59867).

Comparator

Healthy controls or non-HF post-STEMI patients

Outcome

Differentially expressed genes (DEGs) and predictive machine learning models for STEMI and post-STEMI HFsurrogate

Main Result

Effect estimate: AUC 0.9441

Integrated transcriptomic analysis identified a 5-gene machine learning model for STEMI prediction and highlighted BST1 and ITGAM as potential prognostic biomarkers for post-STEMI heart failure.

Limitations

The post-STEMI HF cohort is relatively small.
The databases of gene ontology will be revised such that the analysis may have to be repeated.
The results from the gene array and bioinformatic analysis require further biological proof-of-concept studies to verify.
Large-scale and prospective investigations are required to confirm the clinical feasibility of the proposed biomarkers.
The post-STEMI HF cohort is relatively small
Databases of gene ontology will be revised such that analysis may have to be repeated
Results from the gene array and bioinformatic analysis require further biological proof-of-concept studies to verify

Cite This Study

Xu et al. (2020) conducted an observational in ST-segment elevation myocardial infarction (STEMI) and post-STEMI heart failure. 5-gene machine learning prediction model (SLC2A3, CLEC4D, GPR97, PLAUR, BST1) vs. Healthy controls was evaluated on Prediction of STEMI (combination set) (AUC 0.9441). A machine learning model built with five hub genes (SLC2A3, CLEC4D, GPR97, PLAUR, and BST1) predicted STEMI with an AUC of 0.9441, while ITGAM and BST1 emerged as prognostic biomarkers for post-STEMI HF.

synapsesocial.com/papers/6a10db07cfa01e990d9fac3d https://doi.org/10.21203/rs.3.rs-118025/v1

View Full Paper

Ask AI

Bookmark

December 10, 2020Open Access

Integrated Gene Expression Profiling Analysis Reveals Potential Molecular Mechanisms and Candidate Biomarkers for Early Risk Stratification and Prediction of STEMI and Post-STEMI Heart Failure Patients

View Full Paper

Ask AI

Bookmark

Key result

Five-gene machine learning model predicts STEMI with ~0.94 AUC and identifies post-STEMI HF biomarkers.

AUC 0.9441

Why the study?

To explore the molecular mechanisms and identify candidate whole-blood predictive and prognostic biomarkers in patients with STEMI and post-STEMI heart failure.

Do specific peripheral blood gene expression profiles predict the occurrence of STEMI and subsequent post-STEMI heart failure?

Population

Whole blood gene expression datasets GSE60993, GSE61144, GSE66360, and GSE59867 from NCBI-GEO

Design

Integrated bioinformatics and machine learning analysis

Authors

JXJing Xu YYYuejing Yang

Discussion

Member takes

Overview

May support blood-based biomarkers for STEMI risk; leaves open clinical adoption pending prospective validation.

Study Design

Type

Observational

Structured PICO

Do specific peripheral blood gene expression profiles predict the occurrence of STEMI and subsequent post-STEMI heart failure?

Population

Microarray data profiles of STEMI patients, post-STEMI heart failure (HF) patients, and healthy controls from four GEO datasets (GSE60993, GSE61144, GSE66360, GSE59867).

Comparator

Healthy controls or non-HF post-STEMI patients

Outcome

Differentially expressed genes (DEGs) and predictive machine learning models for STEMI and post-STEMI HFsurrogate

Main Result

Effect estimate: AUC 0.9441

Integrated transcriptomic analysis identified a 5-gene machine learning model for STEMI prediction and highlighted BST1 and ITGAM as potential prognostic biomarkers for post-STEMI heart failure.

Limitations

The post-STEMI HF cohort is relatively small.
The databases of gene ontology will be revised such that the analysis may have to be repeated.
The results from the gene array and bioinformatic analysis require further biological proof-of-concept studies to verify.
Large-scale and prospective investigations are required to confirm the clinical feasibility of the proposed biomarkers.
The post-STEMI HF cohort is relatively small
Databases of gene ontology will be revised such that analysis may have to be repeated
Results from the gene array and bioinformatic analysis require further biological proof-of-concept studies to verify

Cite This Study

synapsesocial.com/papers/6a10db07cfa01e990d9fac3d https://doi.org/10.21203/rs.3.rs-118025/v1

Also Consider

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

View Full Paper

Ask AI

Bookmark