James Y. Dai and Ingo Ruczinski contributed equally to this work.


Missing single nucleotide polymorphisms (SNPs) are quite common in genetic association studies. Subjects with missing SNPs are often discarded in analyses, which may seriously undermine the inference of SNP-disease association. In this article, we compare two haplotype-based imputation approaches and one regression tree-based imputation approach for association studies. The goal is to assess the imputation accuracy, and to evaluate the impact of imputation on parameter estimation. Haplotype-based approaches build on haplotype reconstruction by the expectation-maximization (EM) algorithm or a weighted EM (WEM) algorithm, depending on whether case-control status is taken into account. The tree-based approach uses a Gibbs sampler to iteratively sample from a full conditional distribution, which is obtained from the classification and regression tree (CART) algorithm. We employ a standard multiple imputation procedure to account for the uncertainty of imputation. We apply the methods to simulated data as well as a case-control study on developmental dyslexia. Our results suggest that imputation generally improves over the standard practice of ignoring missing data in terms of bias and efficiency. The haplotype-based approaches slightly outperform the tree-based approach when there are a small number of SNPs in linkage disequilibrium (LD), but the latter has a computational advantage. Finally, we demonstrate that utilizing the disease status in imputation helps to reduce the bias in the subsequent parameter estimation.



Included in

Biostatistics Commons