Adjusting Family Relatedness in Data-driven Burden Test of Rare Variants. Q. Zhang, L. Wang, I. B. Borecki, M. A. Province Division od Statistical Genomics, Washington University School of Medicine, St Louis, MO.
Family data represents a rich resource for detecting association between rare variants (RVs) and human complex traits. However, most RV association analysis methods developed in recent years are data-driven burden tests which can adaptively learn weights for individual variants from observed data, but require permutation to estimate significance, thus are not readily applicable to family data because random permutation will destroy family structure. Direct application of these methods to family data will usually result in a significant inflation of false positives. To overcome this issue, we have developed a generalized, weighted sum mixed model (WSMM) and corresponding computational techniques that incorporate family information into data-driven burden tests, and allow appropriate and efficient mixed-model-based permutation tests in family data. Using simulated and real datasets (including the GAW17 simulated data, the FamHS exome chip data, and the LLFS exome sequence data), we demonstrate that the proposed WSMM method can be used to appropriately adjust for dependence among family members and has a good control of type I errors. We also compare WSMM with famSKAT (the family based version of a widely-used, non-data-driven score test method), showing that WSMM has significantly higher power in some cases (without losing power in most other cases). WSMM provides a flexible analysis framework that accommodates arbitrary family structures of any complexity, and it can be easily extended for binary and time-to-onset traits, and combined with different data-driven burden test methods.
You may contact the first author (during and after the meeting) at