Flexible use of high-density oligonucleotide arrays for single-nucleotide polymorphism discovery and validation

Abstract

A method for identifying and validating single nucleotide polymorphisms (SNPs) with high-density oligonucleotide arrays without the need for locus-specific polymerase chain reactions (PCR) is described in this report. Genomic DNAs were divided into subsets with complexity of ~10 Mb by restriction enzyme digestion and gel-based fragment size resolution, ligated to a common adaptor, and amplified with one primer in a single PCR reaction. As a demonstration of this approach, a total of 124 SNPs were located in 190 kb of genomic sequences distributed across the entire human genome by hybridizing to high-density variant detection arrays (VDA). A set of independent validation experiments was conducted for these SNPs employing bead-based affinity selection followed by hybridization of the affinity-selected SNP-containing fragments to the same VDA that was used to identify the SNPs. A total of 98.7% (74/75) of these SNPs were confirmed using both DNA dideoxynucleotide sequencing and the VDA methodologies. With flexible sample preparation, high-density oligonucleotide arrays can be tailored for even larger scale genome-wide SNP discovery as well as validation.

Figure 1

Schemes of (A) single nucleotide polymorphism (SNP) discovery and (B) validation. (A) Genomic DNA was digested with EcoRI and separated by size on agarose gel. A 250–350 bp fraction was cut from the gel, extracted, and ligated to an adaptor. The ligated DNA was amplified with one primer that interrogated the adaptor sequence. SNPs were screened by hybridizins to VDA. (B) Designated targets were enriched from total genomic DNA by oligonucleotides bound to magnetic beads that interrogated the SNP sites. The genomic DNA was digested with EcoRI and ligated with an adaptor. The enriched targets were amplified with the same primer and screened similarly as in A.

Figure 2

SNP screening with variant detection arrays (VDAs). (A) The whole image of 40 pM targets hybridized with VDA. (B) An expanded view of a portion of A. (C) The same region of image as in B except that the chip was hybridized with enriched targets. (D) A single nucleotide polymorphism (SNP) detected and confirmed by alternative sample preparation. From top to bottom, the wild type (base T at the center), heterozygous SNP (bases C and T at the center), and homozygous mutation (base C at the center) are shown.

Figure 3

Suppression of signals of repetitve sequences with human Cot-1 fraction. The variant detection arrays (VDAs) were hybridized with enriched targets without adding Cot-1 (A) and with 30 ng/μL human Cot-1 fraction (B). The bright signals at the top of the image are attributable to unique sequences that were enriched. The dark region below had little intensity because the sequences were not enriched. The lower portion shows the cross-hybridization from repetitive sequences (A) that was diminished by adding human Cot-1 DNA.