Figure 3. Mismatch error pattern correlation between two independent datasets.

Mismatch error patterns observed in ERCC spike reads are correlated between independent data sets for read position 1-6 (panel A) but not for subsequent positions (position 7-20, panel B). For each position the distribution of errors over the 12 mispriming possibilites was determined (4 nucleotides x 3 mispriming options, summing to 100% per position) and plotted between the dandelion RNA-seq ERCC-dataset (see Figure 1) and ERCC RNA-seq data described in Jiang (2011; GSM517062). The correlation in error distributions between the two data sets shows consistency of specific mismatching errors only in the hexamer binding region (panel A).