Fig. 3

Metrics for RNA-seq. a) The sequencing library size; b) the mapping summaries for sequence reads; c) the counting summaries for uniquely mapped reads; d) the ambiguous reads arising from gene overlapping; on average, the percentage of ambiguous reads drops approximately 3.1 % from non-stranded to stranded RNA-seq, and this drop roughly represents the overlapping arising from opposite strands; e) the correlation for gene expression profile among those eight samples; the samples are clearly clustered by sequencing protocol; f) the boxplot of gene expression