Predictive modeling of angiotensin-converting enzyme and its gene-polymorphisms in the occurrence of cerebral small-vessel disease

Abstract

Abstract Cerebral small vessel disease (CSVD) is the big cause of stroke, and there are many causes of CSVD. The aim of this study is to screen the key causes of CSVD and finally explore the association of Angiotensin-converting enzyme (ACE) and its gene polymorphisms with CSVD by constructing a multivariate modeling research method based on Lasso regression. Clinical data were collected from 184 CSVD patients and 120 controls, and then lasso regression was used to select the four most relevant clinical characteristics of CSVD. Then we divided the sample size of the training set and the validation set according to the ratio of 7:3, and used the ROC curve and DCA curve to evaluate the diagnostic and survival value of the prediction results. Finally, serum ACE expression and ACE genotyping were tested by Elisa and PCR. Four characteristic variables were selected by lasso regression, including age, sex, serum ACE concentration and ACE genotyping. ROC diagnostic curve showed that the AUC value of the validation set was 0.98, which had high diagnostic value. The subsequent DCA curve also showed that these four characteristic variables had a close clinical correlation with CSVD. The final results also confirmed that the serum ACE value of CSVD patients was higher than that of the Control group (p < 0.001), and the gene frequency (D\I = 291\79) in the CSVD group. And Control group gene frequency (D\I = 46\194) There were also some differences (p < 0.0001). The expression of ACE and its genotype, age, and gender contribute to CSVD.