Supplementary Figure 4: Gene co-expression modules in spMNs correlate with sALS disease status (see Fig. 5).

(a) Principal component analysis of 15,614 mRNA transcripts from sALS (n = 12) and control (n = 10) spMN samples illustrates the major features that define the samples with respect to one another. The y-axis depicts the coordinate along each principal component for each sample. The percent contribution to the total variance of the data by each principal component is shown along the x-axis. In order to reduce obscuration, data points are jittered randomly along the x-axis within each bin. (b) Regression analysis of linear models generated by using 6 sample traits as predictors and coordinates along 21 principal components as dependent variables. Heat map indicates F-statistic. The nominal P-value threshold is indicated. No P-values passed Bonferroni-corrected significance. Sample coordinates along PC1 are best explained by sALS disease status, therefore PC1 is referred to as the sALS component. (c) WGCNA clustered 15,614 genes across sALS and control spMNs (n = 22 samples) based on similar network topology. Height metric on dendrogram indicates topological overlap (TO) distance between genes. A dynamic tree-cutting algorithm grouped tightly networked genes, illustrated as low hanging branches, into 52 modules represented by arbitrary colors directly below the branches. Genes falling onto the predominant light grey color are not classified into any module. (d) Summary expression values for each of the 52 modules, known as module eigengenes, are correlated using Pearson’s method in a pair-wise manner to external traits of the samples, such as the sex, sALS disease status, site of ALS onset, age of tissue donor, disease course, or post mortem interval, or correlated to the principal component coordinate of each sample along PC1 (the sALS component). n = number of samples for which there is data for the indicated sample trait, and thus used in the correlation. Colors on heat map indicate Pearson’s pairwise correlation between module eigengene and sample trait. For PC1, correlations outlined in black denote values with a Bonferroni-corrected P-value < 0.01, and these modules were kept for subsequent GO analysis. For sALS disease status, Benjamini-Hochberg-corrected P-values < 0.05 are outlined in black. Some modules uniquely correlate or anti-correlate with a sample trait, while others correlate to more than one sample trait. This data is also shown in Fig. 5b. For additional information, see Supplementary Tables 5a (for linear expression values of sALS data set), 1c (for linear regression test statistics), 5b (for full lists of significantly enriched gene sets and P-values), and 5c (for module assignments and properties).