Fig. 5.

Splicing correlation network in CRC. (a) DEAS evens correlation network of splicing factors. The correlation analyses between expression levels of these 71 splicing factors and the PSI values of each DEAS event were performed in the CRC cohort with splicing regulatory network built among the significant correlations. The shape, size and color of node respectively represent a type (AS event or splicing factor), the value of logFC and change pattern (upregulated or downregulated). The breadth of the line represents the extent of interaction strength. (b-g) Representative dot plots of correlations between expression of splicing factors and PSI values of AS events. Spearman's rank correlation analysis was used for non-normal distribution data (Fig. 3d and e). Person correlation was used for continuous variables that meet normal distribution (Fig. 3b, c, f and g). All the absolute value of r coefficient > 0.5 and all P < .0001.