HIGH THROUGHPUT SNP IDENTIFICATION. To date, LNG has discovered over 100 single nucleotide polymorphisms (SNPs) in more than 50 neuro-genetic candidate genes for addictive behaviors, including alcoholism, & related behaviors. Approximately 20% of the SNPs discovered by LNG result in non-synonimous amino acid changes that may alter function of the encoded protein or alterations in expression of the gene. Genes for SNP screening were selected on the basis of postulated roles in alcohol, treatment response, identification in whole genome or candidate linkage or association study, and the availability of genomic sequence data. Because a goal of the project is to determine the role os SMPs in complex genetic disorders, we focused our screening in efforts on protein coding portions of the candidate gene and possible regulatory regions within and flanking the protein coding regions. We used denaturing high performance liquid chromatography (dHPLC), which detects greater than 90% of sequence variants in DNA duplexes of 150-450 bp in size, to screen a DNA panel composed of 477 genomic DNAs enriched for clinical and ethnic diversiy. The exact makeup of the screening panel was carefully constructed in order to achieve an accurate estimation of allele prequencies for each novel SNP within our primary linkage data sets. The data sets are composed of native American and Caucasian populations. Sequence variants obtained from the primary screening method were characterized and confirmed using gel-based, semi-automated, or capillary-based, fully automated, DNA sequencing methods. The continuing requirement for SNP detection using accurate, high throughput, and lower cost methods motivated us to develop other methods. Our experimental approach took advantage of thermal melting differences between double stranded DNA molecules that carry either a complete sequence match or a single nucleotide mismatch, as found in heterozygous individuals for a particular SNP. Denaturation and re-annealing of the strands in a DNA mixture from an individual that is heterozygous for a SNP will contain at most 50% of the re-annealed DNAs as perfectly matched sequences, while at least 50% of the re-annealed DNAs will have a single base mismatch. The thermal melting characteristics of DNA duplexes with a single base mismatch are inherently less stable than perfectly matched double stranded DNAs. A Perkin Elmer 7700 Sequence Detector was used to monitor the change in fluorescence signal over the course of double strand DNA melting. With this particular instrument, it was possible to perform thermal denaturation experiments & analyze the DNA melting data in 96 samples at one time. We transformed the raw fluorescence readings into first derivative plots that produced a melting profile for each DNA duplex. With these melting profiles, we were able to detect single-base mismatches in double stranded DNAs 100-150 bp in size. We think that this approach is highly suited for large-scale detection of new sequence variants. HIGH THROUGHPUT GENOTYPING. A central issue in high throughput genotyping procedures is accuracy, flexibility, and cost. Because of its assay design flexibility, low error rate, and potential for performing 96 to 384 assays simultaneously, we selected the 5 prime nuclease assay using SNPs either discovered in the LNG or SNPs that have been described by other laboratories for allele frequency determinations in our study populations. The principle of the 5 prime nuclease asay is that two allele-specific detection probes are designed such that they bind to their cognate allele target at the annealing temperature of the DNA amplification primers. During each amplification cycle, the 5 prime exonuclease activity of Taq polymerase cleaves the quenching dye from the allele-specific detection probe, allowing that the allele-specific signal dyd (VIC or FAM) to fluoresce. Genotypes are produced automatically from the increase in fluorescence intensity specific for each allele. The 5 prime nuclease assay is performed on the ABI 7700 & ABI 7900 that have sample throughput capacities of 96 & 384 samples, respecitvely. Using this approach, LNG has performed greater than 6,000 genotypes in the past year. Genotyping error rates have been calculated from triplicate samples & are less than 0.5%. Reagent costs are currently $0.80 per assay. We have reduced cost to approximately $0.50 per assay be developing in-house reagents.
Showing the most recent 10 out of 79 publications
