Library characteristics, ERCC quantification, and coverage. Quality control plots for a stranded ENCODE RNA-seq library of K562 cell Poly-A+ RNA with ERCC spike-ins (library 7). (A) Mismatch rate along reads mapped to all ERRC RNAs. The first 6 bp correspond to the random reverse transcription hexamer-priming site. (B) Scatter plot for sense and antisense read counts per ERCC. (C,D) Scatter plots of read counts versus mass (concentration times length) per ERCC: (C) 100% ERCC library (library 6) and (D) pool of 44 2% ERCC spike-in H. sapiens libraries (libraries 7–50). ERCC-00073 showed aberrant abundance patterns in multiple RNA-seq experiments, as did ERCC-00144 in ERCC pool 14. They may have been inaccurately quantified in our ERCC test set due to errors during the complex mixing scheme used to generate the pools, as they are also suspect in RT-PCR and array experiments on these ERCC pools (M Salit, unpubl.). (E) Scatter plot of read counts in the 100% ERCC library versus a 1% ERCC spike-in D. melanogaster library (library 5). (F) Average sequencing depth and percentage of primary sequence covered for all ERCC transcripts for real data (black) and simulated data (gray).
