Comparison of PrASE and Pyrosequencing for SNP Genotyping

Abstract

Background: There is an imperative need for SNP genotyping technologies that are cost-effective per sample with retained high accuracy, throughput and flexibility. We have developed a microarray-based technique and compared it to Pyrosequencing. In the protease-mediated allele-specific extension (PrASE), the protease constrains the elongation reaction and thus prevents incorrect nucleotide incorporation to mismatched 3'-termini primers.

Results: The assay is automated for 48 genotyping reactions in parallel followed by a tag-microarray detection system. A script automatically visualizes the results in cluster diagrams and assigns the genotypes. Ten polymorphic positions suggested as prothrombotic genetic variations were analyzed with Pyrosequencing and PrASE technologies in 442 samples and 99.8 % concordance was achieved. In addition to accuracy, the robustness and reproducibility of the technique has been investigated.

Conclusion: The results of this study strongly indicate that the PrASE technology can offer significant improvements in terms of accuracy and robustness and thereof increased number of typeable SNPs.

Figure 1

SNP typing by PrASE. A nested multiplex PCR is performed to amplify all SNP loci in a single reaction. The biotin-labeled inner PCR products are captured by streptavidin-coated magnetic beads facilitating automated reaction clean-ups between all assay steps. Strand-specific alkali elution is then performed before hybridization of allele-specific extension primers which contain unique tag sequences for later microarray detection. The multiplex PrASE reaction is performed with Cy5-labeled dNTPs to facilitate fluorescence detection. The products of the reaction are released with alkali, neutralized, and hybridized to a universal tag microarray containing 48 identical wells before detection.

Figure 2

An array-of-microarrays and the corresponding clusters. 48 subarrays are spotted to each glass each having the entire set of 40 anti-tags spotted in triplicates (left panel insert) thus constituting an array-of-arrays (left panel). The colors are artificially added according to signal intensity by the software. The two signals for each SNP are plotted together as cluster diagrams where the three clusters represent each of the three possible genotypes (right panel). Allelic fractions (AFs, x-axes) are calculated from the signal intensities as spot 1/(spot 1 + spot 2) and logarithm of the total signal is used for the y-axes. To investigate the variability between tag sequences, each allele-specific extension primer was designed with two alternative tag sequences. All combinations for each SNP gave similar clusters to the presented.

Figure 3

Robustness of the assay. All the 442 typed samples from all experiments plotted together for each of the SNPs. As illustrated, clusters of each SNP give the same pattern between runs indicating the robustness and consistency of the assay. In fact, the entire dataset can be clustered in one diagram (right panel), still forming three distinct and correct clusters.

Figure 4

Sequences of all PCR amplicons used for PrASE. Inner multiplex PCR primers are underlined. The placement of PrASE and Pyrosequencing primers are indicated in bold and italic respectively. Note that the forward strand sequences are displayed hence the actual sequences of primers are reverse and complementary where appropriate.