Skip to main content
Fig. 5 | BMC Medical Genomics

Fig. 5

From: Revealing Prdx4 as a potential diagnostic and therapeutic target for acute pancreatitis based on machine learning analysis

Fig. 5

Screening of hub genes with diagnostic value via machine learning. (A) 13 diagnostic markers were screened by the least absolute shrinkage and selection operator (LASSO) logistic regression algorithm. (B) 4 diagnostic markers were screened by a support vector machine-recursive feature elimination (SVM-RFE) algorithm. (C) Venn diagram of variables intersecting LASSO and SVM-RFE algorithms. (D) The expression level of hub gene Prdx4 in the merged dataset between AP and control groups. (E) The expression level of hub gene Prdx4 in the in the validation cohort between AP and control groups. (F) Analyze of diagnostic validity of the diagnostic marker Prdx4. (G) ROC validation of diagnostic validity of the diagnostic marker Prdx4 in the validation cohort. AUC = 0.917.

Back to article page