In this application, we request continuation of MH057881, Genetic Association in Schizophrenia and Other Disorders. In our previous aims, covering the last fourteen years, we targeted the development of statistical methods for identifying genetic variants affecting liability to simple and complex disease. Specifically we have developed novel methods to control for population substructure and to fine-map risk variants, targeted methods to characterize linkage disequilibrium (LD) and use LD for association analysis, explored false discovery rate (FDR) procedures for genetic analysis, especially in the setting of high dimensional models, and developed robust methods for common variant associations with disease via genome-wide association. During the next five years, we propose to pursue methods related to association of rare variants with disease status, an area currently of keen interest to human genetics. Next Generation Sequencing (NGS) data have only recently become economically feasible to generate and application of the technology for genetic analysis of complex diseases is underway. As expected NGS data are noisy and signals for association emanating from it are often modest. Nonetheless it is reasonable to expect these data will enhance our understanding of the genetic etiology of complex diseases, but we need good tools to dissect the data. The overarching aim for this R01 is to develop novel statistical methods and evaluate existing methods to extract association signal from NGS data, with particular emphasis on data from disorders affecting mental health. Motivating this proposal we have recently shown the utility of de novo events - mutations in children not found in parents - for identification of genes involved in risk for autism. A portion of our research will build on his foundational observation and how to integrate de novo and inherited variation. Moreover, to garner additional power, we propose to develop methods to incorporate biological information, such as the nature of sequence variation (loss-of-function, missense, and silent) and gene networks, with the distribution of rare variation in subjects to identify disease genes. As has been true for our last three funding periods, our theoretical work will be guided by real data from the evolving field of human genetics. We are well positioned to move between theory and data because we have a diverse team of investigators lead by the PI (Devlin) and subcontract PI (Roeder) who have decades of experience in the statistical genetics field.
Whether diseases are common in the population, such as major depression and heart disease, or relatively uncommon, such as psychiatric disorders, their genetic causes are often obscure. Yet determining the genetic variants underlying these diseases could be a major step toward improving the health and well being of mankind. To accomplish this goal, researchers need the right tools: our research group develops statistical tools to identify risk genes, especially those affecting mental health.
Showing the most recent 10 out of 72 publications
