Disclaimer: Supplementary materials have been peer-reviewed but not copyedited.


Supp. Figure S1. Prevalence of the most frequent KRAS CDS mutations identified in 10,342 colorectal adenocarcinomas in the COSMIC database (

Supp. Figure S2. Validation of the KRAS mutations in the five selected cancer cell lines. A. Representation of the sequences analysed by pyrosequencing in the different cell lines including codon 12 and codon 13 represented in pink and grey, respectively. B. Genotyping of KRAS codon 12 and 13 mutations in the five cancer cell lines which harbour c.38G>A (DLD1), c.34G>A (A549), c.35G>T, (SW480), c.35G>A (LS174T) and c.35G>C (RPMI-8226) mutations.

Supp. Figure S3. Representation of the different standard, ice-COLD- and Enhanced-ice-COLD-PCR assays used for the detection and identification of KRAS mutation in codon 12 and codon 13.

Supp. Figure S4. Tc determination for the KRAS ice-COLD-PCR assay. Tc was determined as the Tm of the smallest peak of 50% mutation of DLD1, LS174T and RPMI-8226 cell line using the 78 bp amplification product. Melting was performed from 65 to 95°C increasing by 0.2°C per step (4 sec hold at each step). The Tm of the WT amplicon was determined as 81°C.

Supp. Figure S5. Development of an automated MS Excel Visual Basic application for KRAS mutation detection, quantification and identification from pyrosequencing data. A. Pyrosequencing peak distributions for c.38G>A, c.34G>A, c.34G>T, c.35G>A, c.34G>T and c.34G>C homozygote mutations and WT. 0, 1 and 2 indicate the absence of nucleotide peaks, the presence of one and two nucleotide peaks respectively. B. Formulas used for the different mutation percentage quantifications. C. Quality control rules for exclusion of non-compliant peak patterns. D. Graphical user interface of the macro.

Supp. Figure S6. Correlation between the observed and the theoretical percentage of mutant allele in DLD-1 (A), A549 (B), SW480 (C), LS174T (D) and RPMI 8226 (E) for the KRAS 114 assay. Pearson correlation coefficient R2 obtained after standard PCR, ice-COLD-PCR, Enhanced-ice-COLD-PCR I and II (using a blocker overlapping with five bases of the 3’ end of the reverse primer or not overlapping with the reverse primer at all) are shown in the graph.

Supp. Figure S7. Correlation between the observed and the theoretical percentage of mutant allele in DLD-1 (A), A549 (B), SW480 (C), LS174T (D) and RPMI 8226 (E) for the KRAS 96 bp assay. Pearson correlation coefficient R2 obtained after standard PCR or Enhanced-ice-COLD-PCR are shown in the graph.

Supp. Figure S8. Quantification of KRAS mutation enrichment by WTB-PCR, ice-COLD-PCR and Enhanced-ice-COLD-PCR compared to standard PCR using the 114 bp assays. The WTB-PCR and E-ice-COLD-PCR II assay used a 31 nucleotides blocker and differed by cycling conditions. A. Quantification of the mutant allele by pyrosequencing after standard PCR, WTB-PCR, ice-COLD-PCR or Enhanced-ice-COLD-PCR. B. Quantification of the percentage of mutation enrichment after WTB-PCR, ice-COLD-PCR or E-ice-COLD-PCR compared to a standard PCR. C. Calculation of fold enrichment of the mutant allele after WTB-PCR, ice- COLD-PCR or Enhanced-ice-COLD-PCR. Numbers after each cell line indicate the percentage of mutant allele.

Supp. Figure S9. Sanger sequencing from the fresh frozen cohort of eight CRC samples identified as mutated (6, 8, 36, 56, 64, 80, 84, 94 & 100), of one non-mutated CRC sample (11) and a WT sample after E-ice-COLD-PCR and Standard PCR. WT KRAS sequence is shown below the chromatograms and codons 12 and 13 are underlined. Mutated nucleotides are indicated by an arrow.

Supp. Table S1. List of primers and probes used in the study

Supp. Table S2. Quantification of the c.34G>A mutation by pyrosequencing after E-ice-COLD-PCR using a gradient of the critical temperature (Tc) and blocker concentrations on a 0.5% dilution of the A459 cell line dilution for determination of the optimal Tc of the KRAS 96 bp assay

Supp. Table S3. Determination of analytical sensitivity (limit of detection) of the three E-ice-COLD-PCR assays tested for five mutations by pyrosequencing

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.