Figure 2
From: RNA-Seq of human whole blood: Evaluation of globin RNA depletion on Ribo-Zero library method
Read distributions across methods and samples. (a) Estimated fraction of reads that map to hemoglobin (Hgb) genes, rRNA, and Other in method-paired datasets from 6 human donors (Experiment 1) and technical replicate datasets from C1 and C2 donors (Experiment 2) with either Globin-Zero (GZ) or Ribo-Zero-Gold (RZG). Proportions were estimated by mapping a subset of total trimmed reads to only Hgb gene sequences or only rRNA gene sequences, intentionally including multi-mapped reads. (b) HgbRNA levels in C1 and C2 libraries. Counts per million (CPM) expression levels measured in Experiment 2 GZ and RZG replicate libraries for Hgb genes were used to calculate the average log2 CPM for each gene in the C1 and C2 samples with each library method. Hgb gene family members with fewer than 20 raw counts in all libraries are not shown. (c) Ensembl gene biotype (https://uswest.ensembl.org/info/genome/genebuild/biotypeshtml.)Â proportions from Experiment 2 utilizing the raw gene counts (number of uniquely-mapped reads that unambiguously map to only one gene) that were used for subsequent quantitative analysis. Technical replicates are averaged. RZG-Hg indicates gene counts from RZG libraries after bioinformatic removal of Hgb gene counts. (d) Gene biotype proportions of counts after TPM calculation which estimates the number of RNA transcripts from each gene by not only normalizing the raw gene counts by library depth as CPM does but also by accounting for gene length; this permits count comparison of differently-sized transcripts within a sample. Technical replicates are averaged.
