?Statistical methods and analysis to detect pleiotropy in GWAS of oral cleft trios? Orofacial clefts (OFCs) are one of the most common human craniofacial birth defects that are broadly categorized as cleft lip with or without cleft palate (CL/P) and cleft palate (CP) alone based on epidemiologic and embryologic patterns. OFCs affect ~1/1000 live births worldwide although the prevalence differs by ancestry and geography. They have a complex etiology and exhibit strong familial aggregation. While at least a dozen different genes are recognized to influence risk of CL/P, very few are known to affect risk of CP and only 1 gene confirmed to affect risk of both. Biological mechanisms controlling risk to OFCs remain largely unknown. Identifying shared genetic risk factors of the OFC subtypes can elucidate genetic architecture of OFCs. Further, identifying shared genetic control of congenital malformations may also improve understanding of OFC genetics because birth defects may share common biological pathways controlling fetal development that can manifest as different anatomical malformations. Case-parent trio design is typically used to collect multi-ethnic samples with OFCs (or other congenital malformations) due to its many advantages in genetic studies of rare disorders. However, no statistical method can appropriately test if a genetic factor has a significant effect on multiple qualitative traits when one or the other qualitative trait is used to independently ascertain the trios. In this R03 application, we propose to develop new, more powerful, cross-phenotype statistical approaches and use them to identify genetic risk factors common to two or more traits using existing genome-wide data on OFC trios. While the methods development is motivated by OFC GWAS data on trios, they can be more broadly applied to identify genetic regions with possible pleiotropic effects on other distinct traits and can be extended to other types of genetic data, including -omics data.
In Aim 1, we will develop a new method based on GWAS summary statistics to test pleiotropy in two traits used to ascertain trios (e.g. OFC subtypes).
In Aim 2, we will develop and support free software to implement our method.
In Aim 3, we will use our method to analyze existing genome-wide data on trios from the POFC and the GENEVA studies to identify genetic risk factors shared by both CL/P and CP in an effort to advance knowledge of OFC genetics.
In Aim 4, we will extend our method in Aim 1 to support >2 traits from family studies, which may then be used to identify shared genetic signals across multiple birth defects including OFCs, thereby improving understanding of OFC genetics. Successful completion of these aims will have immediate impact on OFC investigation; methods & software will be useful to research groups in the Kids First initiative and to the general scientific community due to the broad applicability of our statistical methods.
Orofacial clefts (OFCs) are one of the most common human craniofacial birth defects that are broadly categorized as cleft lip with or without cleft palate (CL/P) and cleft palate (CP) alone based on epidemiologic and embryologic patterns. They have a complex etiology and exhibit strong familial aggregation, and recognized risk genes for OFCs can explain only a small fraction of the estimated heritability of OFCs. In this R03 application, we propose to develop novel, powerful and first-of-its-kind statistical methods for identifying potential pleiotropic genes affecting risk of two or more traits, especially qualitative traits where one or the other trait is used to independently ascertain case-parent trios; develop and support a free software to implement our methods; and apply our methods to analyze existing genome-wide data on multi-ethnic, case-parent trios (child- mother-father ascertained through OFCs) from the POFC and GENEVA studies to identify shared genetic signals between CL/P and CP, thereby advancing knowledge of OFC genetics.
