1 Introduction

The SnipSnip program is the implementation of a GWAS method to detect causal variants using poorly tagged data by using multiple SNPs in low LD with the causal variant. Genome-wide association studies (GWAS) allow the detection of non-genotyped disease causing variants through the testing of nearby genotyped SNPs that are in strong linkage disequilibrium (LD) with the causal variant. This approach is naturally flawed when there are no genotyped SNPs in strong LD with the causal variant. There may, however, be several genotyped SNPs in weak LD with the causal variant that, when considered together, provide equivalent information. This observation provides the motivation for the popular (but computationally intensive) imputation-based approaches that are often used.

SnipSnip is designed for the scenario where there are several genotyped SNPs in weak LD with a causal variant. Our approach proceeds by selecting, for each genotyped “anchor” SNP, a nearby genotyped “partner” SNP (chosen, on the basis of a specific algorithm we have developed, to be the optimal partner). These two SNPs are then used as predictors in a linear or logistic regression analysis, in order to generate a final significance test associated with the anchor SNP.

SnipSnip is designed for use with unrelated individuals either in a case-control analysis or a quantitative trait analysis.

Our method, in some cases, potentially eliminates the need for more complex methods such as sequencing and imputation or haplotype analysis, and provides a useful additional test that may be used with existing GWAS data to identify genetic regions of interest.

For an example application of SnipSnip see Chen et al. (2015).

1.1 Program information and citation

For details concerning the methodology of the SnipSnip GWAS, please see the accompanying manuscript Howey and Cordell (2014).

The program SnipSnip is written in C++ and executables are available for Linux and Windows from the download page, as well as the source code.

next->