Fig. 5 | Scientific Reports

Fig. 5

From: Identification of palmitoylated biomarkers in non-alcoholic fatty liver disease via integrated bioinformatics analysis and machine learning

Fig. 5

Evaluation of the diagnostic potential of candidate biomarkers and construction of a nomogram in the training (GSE89632) and validation cohorts (GSE164760). (A) The violin plots elucidate the differential expression profiles of TYMS, WNT5A, and ZFP36 between the control and the NAFLD group in the training cohort. (B) A predictive nomogram for NAFLD occurrence constructed from PRHGs. “Points” indicate the scores of hub genes, and “Total Points” represent the sum of all these gene scores. (C) The calibration curve illustrates the relationship between actual and predicted NAFLD rates. A perfectly accurate model would be represented by the dotted diagonal line. The performance of the nomogram is depicted by the solid line, with its closeness to the diagonal indicating greater precision. (D) ROC curve of the nomogram in training cohort for NAFLD. (E) The violin plots elucidate the differential expression profiles of TYMS, WNT5A, and ZFP36 between the control and the NAFLD group in the validation cohort. (F) A predictive nomogram for NAFLD occurrence constructed from PRHGs in the validation cohort. (G) The ROC curves and corresponding AUC values for TYMS, WNT5A, and ZFP36 in the validation cohort. (H) ROC curve of the nomogram in validation cohort for NAFLD.

Back to article page