FAM171B as a Novel Biomarker Mediates Tissue Immune Microenvironment in Pulmonary Arterial Hypertension

Figure 7

Construct multiple machine learning models based expression of DEGs. (a) The effect of the decision tree number on the error rate. The -axis denotes the number of decision trees, while the -axis shows the error rate. When approximately 100 decision trees are used, the error rate is generally steady. (b) The results of Gini coefficient method in a random forest classifier. The -axis displays the genetic variable, and the -axis the significance index. (c) Fine-tuning the least absolute shrinkage and selection operator (LASSO) model’s feature selection. LASSO regression was used to narrow down the DEGs, resulting in the discovery of 28 variables as potential markers for PAH. The ordinate represents the value of the coefficient, the lower abscissa represents log(λ), and the upper abscissa represents the current number of nonzero coefficients in the model. (d) A plot illustrating the process of selecting biomarkers using the support vector machine-recursive feature elimination (SVM-RFE) technique. The SVM-RFE technique was used to identify a subset of 37 characteristics from the DEGs. DEGs: differentially expressed genes; PAH: pulmonary arterial hypertension.

(a)

(b)

(c)

(d)