High-throughput genotyping of single nucleotide polymorphisms using new biplex invader technology

Abstract

The feasibility of large-scale genome-wide association studies of complex human disorders depends on the availability of accurate and efficient genotyping methods for single nucleotide polymorphisms (SNPs). We describe a new platform of the invader assay, a biplex assay, where both alleles are interrogated in a single reaction tube. The assay was evaluated on over 50 different SNPs, with over 20 SNPs genotyped in study cohorts of over 1500 individuals. We assessed the usefulness of the new platform in high-throughput genotyping and compared its accuracy to genotyping results obtained by the traditional monoplex invader assay, TaqMan genotyping and sequencing data. We present representative data for two SNPs in different genes (CD36 and protein tyrosine phosphatase 1beta) from a study cohort comprising over 1500 individuals with high or low-normal blood pressure. In this high-throughput application, the biplex invader assay is very accurate, with an error rate of <0.3% and a failure rate of 1.64%. The set-up of the assay is highly automated, facilitating the processing of large numbers of samples simultaneously. We present new analysis tools for the assignment of genotypes that further improve genotyping success. The biplex invader assay with its automated set-up and analysis offers a new efficient high-throughput genotyping platform that is suitable for association studies in large study cohorts.

Figure 1

Schematic of the invader assay. During the primary phase, an Invader^® oligo and a primary probe are annealed to target DNA, overlapping at the SNP position (indicated in upper case letters). The black arrow indicates the site of cleavage by the cleavase enzyme. The released 5′ flap anneals to the FRET cassette during the secondary phase and initiates a second cleavage reaction that releases the fluorescent dye. The signal is only released when the invasive structure is formed on the target DNA (‘Perfect Match’, left reaction). If the primary probe does not match the nucleotide at the SNP position, cleavase will not act (reaction on right).

Figure 2

Sector definition for CA clustering algorithm. The initial four cluster areas defined by the heuristic part of the clustering algorithm. The crosses denote the centers of gravity of the clusters (corresponding to areas of high density). Lines connecting the centers indicate how the midpoints are determined. The midpoints connect to the origin to define the three initial sectors.

Figure 3

Scatter plot of fluorescence data from biplex invader assay for PTP03. Raw fluorescence data are plotted for each sample. The x-axis depicts the fluorescence intensity for dye 1, corresponding to SNP allele 1, while the y-axis indicates the fluorescence intensity for the second dye, corresponding to the alternative SNP allele. Four clusters can be identified, one consisting primarily of no-DNA control samples and samples that failed to generate signal (0/0) and the remaining three clusters indicating the three possible genotypes (homozygous for allele 1, 1/1; homozygous for allele 2, 2/2; heterozygous, 1/2).