SNP uniqueness problem: a proof-of-principle in HapMap SNPs

Area of Science:

• Genomics
• Bioinformatics
• Medical Genetics

Background:

• Single nucleotide polymorphisms (SNPs) are crucial for biomedical and clinical research.
• The presence of nonunique or false-positive SNPs in datasets can bias research findings and lead to inaccurate conclusions.
• The HapMap dataset is a widely used resource for SNP information.

Purpose of the Study:

• To computationally assess the extent of nonunique and false-positive SNPs within the HapMap dataset.
• To evaluate the impact of nonunique SNPs on clinical association studies and genotyping arrays.

Main Methods:

• Utilized BLAT analysis with two sets of SNP flanking sequences against the human genome.
• Assessed the representation of identified nonunique SNPs in commercial genotyping arrays and clinical association databases.

Main Results:

• 4.2% and 11.9% of HapMap SNPs showed nonunique alignment to the human genome (long and short sequences, respectively).
• An average of 7.9% of nonunique SNPs were found in common commercial genotyping arrays.
• Identified nonunique SNPs are present in clinical association databases, indicating potential inaccuracies.

Conclusions:

• A significant proportion of SNPs in the HapMap dataset are nonunique, potentially compromising the accuracy of genetic research.
• The findings raise concerns about the validity of certain disease-related genotyping analyses due to SNP annotation errors.