A 2 miRNAs-based signature for the diagnosis of atherosclerosis
| Autor: | Huimin Wang, Xiujiang Han, Lina Liu, Sheng Gao, Yongjian Li |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
lcsh:Diseases of the circulatory (Cardiovascular) system
Logistic regression model Computational biology 030204 cardiovascular system & hematology Logistic regression Signature Risk Assessment Correlation 03 medical and health sciences 0302 clinical medicine Predictive Value of Tests Risk Factors microRNA Diagnosis Medicine Humans In patient Gene 030304 developmental biology Oligonucleotide Array Sequence Analysis Gene expression omnibus 0303 health sciences Microarray analysis techniques business.industry WGCNA Gene Expression Profiling Stepwise regression Atherosclerosis MicroRNAs lcsh:RC666-701 Case-Control Studies miRNAs Cardiology and Cardiovascular Medicine business Databases Nucleic Acid Transcriptome Research Article |
| Zdroj: | BMC Cardiovascular Disorders BMC Cardiovascular Disorders, Vol 21, Iss 1, Pp 1-8 (2021) |
| ISSN: | 1471-2261 |
| Popis: | Background Atherosclerosis (AS) is a leading cause of vascular disease worldwide. MicroRNAs (miRNAs) play an essential role in the development of AS. However, the miRNAs-based biomarkers for the diagnosis of AS are still limited. Here, we aimed to identify the miRNAs significantly related to AS and construct the predicting model based on these miRNAs for distinguishing the AS patients from healthy cases. Methods The miRNA and mRNA expression microarray data of blood samples from patients with AS and healthy cases were obtained from the GSE59421 and GSE20129 of Gene Expression Omnibus (GEO) database, respectively. Weighted Gene Co-expression Network Analysis (WGCNA) was performed to evaluate the correlation of the miRNAs and mRNAs with AS and identify the miRNAs and mRNAs significantly associated with AS. The potentially critical miRNAs were further optimized by functional enrichment analysis. The logistic regression models were constructed based on these optimized miRNAs and validated by threefold cross-validation method. Results WGCNA revealed 42 miRNAs and 532 genes significantly correlated with AS. Functional enrichment analysis identified 12 crucial miRNAs in patients with AS. Moreover, 6 miRNAs among the identified 12 miRNAs, were selected using a stepwise regression model, in which four miRNAs, including hsa-miR-654-5p, hsa-miR-409-3p, hsa-miR-485-5p and hsa-miR-654-3p, were further identified through multivariate regression analysis. The threefold cross-validation method showed that the AUC of logistic regression model based on the four miRNAs was 0.7308, 0.8258, and 0.7483, respectively, with an average AUC of 0.7683. Conclusion We identified a total of four miRNAs, including hsa-miR-654-5p and hsa-miR-409-3p, are identified as the potentially critical biomarkers for AS. The logistic regression model based on the identified 2 miRNAs could reliably distinguish the patients with AS from normal cases. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink