Powerful and Adaptive Testing for Multi-trait and Multi-SNP Associations with GWAS and Sequencing Data

• DOI: 10.1534/genetics.115.186502
• Publication Date: 2016-04-14T05:39:22Z
• Authors: Junghi Kim; Yiwei Zhang; Wei Pan
• Keywords: Univariate; Trait; SNP; Genome-wide Association Study; Statistical power; Multiple comparisons problem; Genetic Association; Association test; Genetic architecture

Testing for genetic association with multiple traits has become increasingly important, not only because of its potential to boost statistical power, but also direct relevance applications. For example, there is accumulating evidence showing that some complex neurodegenerative and psychiatric diseases like Alzheimer's disease are due disrupted brain networks, which it would be natural identify variants associated a network, represented as set traits, one each regions interest. In spite promise, testing multivariate trait associations challenging: if appropriately used, power can much lower than on univariate separately (with proper control testing). Furthermore, differing from most existing methods single-SNP-multiple-trait associations, we consider SNP set-based decipher complicated joint effects SNPs traits. Because the test critically depends several unknown factors such proportions propose highly adaptive at both levels, giving higher weights those likely yield high across wide spectrum situations. We illuminate relationships among proposed tests, covers tests special cases. compare performance new using simulated real data. The were applied structural magnetic resonance imaging data drawn Disease Neuroimaging Initiative genes gray matter atrophy in human default mode network (DMN). genome-wide studies (GWAS), AMOTL1 chromosome 11 APOE 19 discovered by significantly DMN. Notably, gene was detected single SNP-based analyses. To our knowledge, been highlighted other before, although indicated related cognitive impairment. method applicable rare sequencing extended pathway analysis.