Fig. 3
Exploratory analysis of RNA-seq samples similarity. Clustering and principal component analysis show that samples are grouping by biological replicates and experimental design. There exists a sharp difference between NExp and DSLT groups while the samples split into two subgroups inside DSLT depending on whether a culture was exposed to light. The sample pairwise distances were computed on a matrix of regularized log-transformed transcript counts. (a) Heat map of sample distances. Hierarchical sample clustering is presented on the top and left sides. (b) Principal component analysis of RNA-seq samples.
