Fig 1.
Flow chart for the whole study.
Fig 2.
Weighted correlation network analysis.
(A) Analysis of scale-free fit parameter and mean connectivity for various soft-thresholding powers. (B) Cluster dendrogram of genes, with dissimilarity based on topological overlap. (C) Co-expression similarity of entire modules based on hierarchical clustering of module eigengenes and the correlation between different modules. The color of cells in the heatmap represented the correlation coefficients of different sizes. Specifically, red colors stood for the positive correlations and blue colors represented the negative correlations. (D) Distribution of average gene significance and errors in the modules associated with the response to CPAP therapy. (E) Correlation heat map of gene modules and phenotypes. The figure without brackets in each cell indicated the clinical feature correlation coefficients. The corresponding p-value was shown below in parentheses. Red is positively correlated with the phenotype; green is negatively correlated with the phenotype.
Table 1.
Top 15 GO enrichment terms associated with genes in black module.
Table 2.
Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis of genes in black module.
Fig 3.
Identification of hub genes by intersection of black module genes and top 200 DEGs.
(A) Heat map of the top 200 DEGs. (B) Venn plot of common genes.
Fig 4.
Construction of random forest model.
(A) The scatter plot of the false-positive rate. The vertical axis represents the false positive rate, and the horizontal axis represents mtry index (from 1 to 11). (B) Relationship between the related errors and the number of decision trees. The vertical axis represents the related errors, and the horizontal axis represents the number of decision trees. (C) The scatter plot of the random forest variable based on the index of mean decrease accuracy (the left panel) and mean decrease Gini (the right panel), respectively. (D) Evaluation of the prediction efficiency of the random forest model via ROC curve. Red line represents the random forest model, and green line represents OBP2B gene.
Fig 5.
Evaluation and visualization of immune cell infiltration.
(A) Stacked barplot diagram of immune cell types of 30 samples in GSE133601. (B) Boxplot diagram of immune cells. (C) Correlation heat map of immune cells. Blue represents a positive correlation, red represents a negative correlation. The darker the color, the stronger the correlation.