ARMS detected an additional 32 mutations. Eighteen of these were not detected on the sequencing traces and 14 failed to sequence. Three mutations were detected by sequencing only. These were mutations that the ARMS assays were not designed to detect. (B) NSCLC mutations. Eight EGFR mutations were detected in the NSCLC samples by both methods. ARMS detected an additional 10 mutations. Two of these were not analysed

by sequencing as the DNA amount was too low and eight failed to sequence. Nine mutations were detected by sequencing only. These were mutations that the ARMS assays were not designed to detect. Note that there were 27 mutations in 26 patients as one sample was found to contain two mutations. DNA quantity and DMXAA molecular weight ability to detect mutations The first 121 of the melanoma samples yielding DNA were grouped by DNA yield to determine if at low DNA quantity MRT67307 ic50 the ability to detect mutations was reduced. The groupings (>5 copies, 5-9 copies, 10-49 copies, 50-99 copies, 100-500 copies and >500 copies) were based on the amount of DNA in the control reactions that could be used to estimate the amount of DNA in the sample. There were more groupings at the lower DNA concentration as it was thought that any effect would

be more likely to be observed in these samples. There was no decrease in the ability to detect mutations as the DNA amount decreased. Both DNA sequencing and ARMS gave similar results in each category although overall ARMS detected more mutations. As the DNA concentration increased the number of successful sequencing reactions also increased: at >50 copies per assay input, the analysis success rate was very similar for both ARMS and sequencing. The results are shown in Fig. 2. Figure 2 Mutation detection success on varying the amount of input DNA. The DNA yield was grouped into categories and the percentage of mutations detected calculated for each group. The n values are the successful number of sequencing and ARMS analyses. The

lower yielding samples did not show any decrease in the numbers of BRAF or NRAS mutations detected. Both DNA sequencing and ARMS gave similar results in each category although overall ARMS detected more mutations. As the DNA concentration increased the number of successful sequencing reactions also increased: at >50 copies per assay Carnitine palmitoyltransferase II input, the analysis success rate was very similar for both ARMS and sequencing. In some samples at high DNA concentrations (>1000 copies assay input) non-specific signal did occur in the ARMS. In these samples it was important to dilute DNA below 1000 copies per assay input and repeat the analysis. This only affected a minority of samples – most samples in excess of this DNA limit did not exhibit any non-specificity at all. Why this should occur in some samples and not others is not known but adds to the difficulty of analysing FF-PET DNA.