Fig. 8 | Scientific Reports

Fig. 8

From: Multi-modal transcriptomic analysis reveals metabolic dysregulation and immune responses in chronic obstructive pulmonary disease

Fig. 8

Consensus clustering of COPD samples. (A) presents a cumulative distribution function (CDF) figure for consensus clustering, highlighting the relative change in consensus indices from k = 2 to k = 9, with the curve exhibiting the most stable change indicating the optimal number of clusters. (B) Visualizes the clustering trajectory from k = 2 to k = 9. (C) Features a consensus matrix, where distinct blank areas between blue modules signify a successful analysis. (D) Illustrates the principal component analysis (PCA) of COPD samples, with scatter plots delineating feature genes differentiating COPD into two subtypes, C1 and C2. (E) Details a KEGG analysis comparing these subtypes, where purple and green represent up- and down-regulated functional pathways, respectively. (F) Shows a box plot contrasting the expression of feature genes between C1 and C2. Finally, (G) displays a heatmap correlating C1 and C2 with feature genes, using purple to indicate up-regulation and green for down-regulation, with *P < 0.05, **P < 0.01, ***P < 0.001 denoting statistical significance levels.

Back to article page