Additional file 5 of Limitations of alignment-free tools in total RNA-seq quantification

Impact of gene lengths and expression levels on gene quantification correlations among pipelines. (a) Gene lengths influence gene expression estimation correlations. Genes were separated into quantile groups according to their lengths. For every quantile group from each sample, Pearson’s correlation coefficients of average TPMs for each gene across replicates were computed for every pairwise pipeline comparison and then averaged across the four samples. Plotted are the average correlations versus the quantile groups. Each line represents a pairwise comparison and is color-coded accordingly. (b) Gene expression levels influence gene expression estimate correlations. Genes were separated into quantile groups according to their average expression values (TPMs). For every quantile group from each sample, Pearson’s correlations were computed in the same way as in panel (a). Each line represents a pairwise comparison and is color-coded accordingly. (PNG 412 kb