For all three simulations, error rates increased (to varying extents) towards the tail end of the read (Figure 1). While this is a recognised issue for Illumina sequencing, the literature is ambiguous with respect to Roche/454 sequencing data [8,9]. While a positional effect for Roche/454 can be seen in Figure 1, the magnitude of this effect is extremely small when compared to Illumina sequencing data and would probably not be observed if only reads with length < 400 bp are considered. The positional increase in error rate for Illumina v5 is also rather modest, and is only slightly more elevated for the second read in the pair. In contrast, Illumina v4 exhibits a strong influence of base position within the read on error rate, characterised by sudden peaks and a marked increase in error rates near the end of the read.
