Joint Efficacy of the Three Biomarkers SNCA, GYPB and HBG1 for Atrial Fibrillation and Stroke: Analysis Via the Support Vector Machine Neural Network

Atrial fibrillation (AF) is the most common type of persistent arrhythmia. Although its incidence has been increasing, the pathogenesis of AF in stroke remains unclear. In this study, a total of 30 participants were recruited, including 10 controls, 10 patients with AF and 10 patients with AF and stroke (AF + STROKE). Differentially expressed genes (DEGs) were identified, and functional annotation of DEGs, comparative toxicogenomic database analysis associated with cardiovascular diseases, and predictions of miRNAs of hub genes were performed. Using RT-qPCR, biological process and support vector machine neural networks, numerous DEGs were found to be related to AF. HBG1, SNCA and GYPB were found to be upregulated in the AF group. Higher expression of hub genes in AF and AF + STROKE groups was detected via RT-PCR. Upon training the biological process neural network of SNCA and GYPB for HBG1, only small differences were detected. Based on the support vector machine, the predicted value of SNCA and GYPB for HBG1 was 0.9893. Expression of the hub genes of HBG1, SNCA and GYPB might therefore be significantly correlated to AF. These genes are involved in the incidence of AF complicated by stroke, and may serve as targets for early diagnosis and treatment.