sequencing read, we are able to accurately estimate the position-dependent sequence read error rate, which is known to be variable in sequencing by synthesis [6]. This facilitates the sensitive and specific detection of low prevalence mutations in the tumor samples. We designed 676 primer pairs to target 518 cancer mutational hotspots of 42 cancer genes, as well as 158 calibration amplicons (see below; Table S2 in Additional file 2).
