Fig. 2
Identification of key modules associated with LUAD using WGCNA. (a) Selection of the soft threshold power (β) for WGCNA analysis. A soft threshold power of β = 5 was chosen to ensure that the gene interaction network followed a scale-free topology (scale-free R2 = 0.9). (b) Co-expression network analysis identified thirty-eight distinct modules through WGCNA, each representing clusters of co-expressed genes. (c) Correlation analysis between the identified modules and LUAD. The MEpink module exhibited the highest positive correlation with LUAD (correlation coefficient = 0.72, P = 3e-87), while the MEblue module displayed the highest negative correlation with LUAD (correlation coefficient = -0.83, P = 6e-141). (d) Scatterplots illustrating the relationship between gene significance (GS) and module membership (MM) within the MEblue and MEpink modules. Both the MEblue (correlation coefficient = 0.89, P = 1e-200) and MEpink (correlation coefficient = 0.91, P = 1e-200) modules showed strong correlations, highlighting their relevance to LUAD.
