Skip to main content
Fig. 4 | Genome Biology

Fig. 4

From: Gene-level differential analysis at transcript-level resolution

Fig. 4

Analysis of positionally biased RNA-seq data using TCC aggregation. A log-log plot of p values comparing aggregated sleuth-derived TCC p values using the Lancaster method (x-axis) to p values obtained by differential analysis in DESeq2 with gene counts (y-axis) shows overall agreement (a). DESeq2 applied on gene counts discovered 460 DE genes (FDR < 0.05); Lancaster aggregation on TCCs discovered 243 genes (FDR < 0.05). TCC aggregated analysis can detect differential 3’ UTR usage that is masked in gene count analyses (b). An example is shown from the rat gene Tap1, with rectangular blocks representing individual exons (blank = non-coding, solid = coding), and distinct equivalence classes (ECs) labeled with brackets. Two other transcripts and their corresponding (zero count) equivalence classes are not shown. Significance levels for Tap1 under effects of alveolar stretching were calculated using the Lancaster method (p value = 0.0056) and compared to p values derived from gene counts (p value = 0.169)

Back to article page