Fast and robust association tests for untyped SNPs in case-control studies

Area of Science:

• Genetics
• Statistical Genetics
• Bioinformatics

Background:

• Genome-wide association studies (GWAS) genotype millions of single nucleotide polymorphisms (SNPs) to identify genetic variants associated with traits.
• However, analyzing SNPs not directly genotyped (untyped SNPs) is crucial for refining association signals and integrating diverse datasets.
• Existing methods for untyped SNP analysis can be computationally intensive.

Purpose of the Study:

• To develop a novel, computationally efficient statistical method for testing associations with untyped SNPs in case-control GWAS.
• To account for the uncertainty in estimating genotypes at untyped SNPs.
• To provide a faster alternative to existing untyped SNP analysis approaches.

Main Methods:

• Proposed a novel statistical approach utilizing an efficient score function derived from a prospective likelihood.
• Employed computationally efficient phasing methods and haplotype sharing measures to infer untyped SNP genotypes.
• Developed the 'untyped' software package to implement the proposed methodology.

Main Results:

• The novel approach automatically accounts for variability in estimating untyped SNP genotypes.
• The method is computationally significantly faster than existing untyped analysis approaches.
• Simulated data demonstrated performance nearly equivalent to hidden Markov model-based methods.

Conclusions:

• The developed statistical method offers an efficient and accurate way to perform untyped SNP analysis in GWAS.
• This approach enhances the utility of GWAS by enabling the inclusion of a broader range of genetic variants.
• The 'untyped' software package facilitates the application of this advanced analytical technique.