In nature, most functional regions of the genome for a chromosome pair express equally. A variation from this equivalence results in genomic imprinting, a phenomenon also called parent-of-origin effect. Using statistical approaches to map imprinted genes (or imprinted quantitative trait loci (iQTL)) has shown promising results. However, current mapping approaches based on diploid genome are limited when the study population is polyploidy such as triploid endosperm. Moreover, current mapping approaches by analyzing phenotypic data measured at a single time point are too simple to take into account developmental or dynamic imprinting trajectories. In this project the investigator develops a collection of novel statistical models for mapping imprinted genes that govern phenotypic traits of interests. The objectives of the proposed research are (a) to have a thorough statistical investigation of the variance component model for mapping imprinted genes underlying single or multiple endosperm traits; (b) to propose methodologies for modeling imprinted gene interactions; (c) to develop functional iQTL mapping approaches for mapping of complex dynamic imprinted traits. With the development of human HapMap project and the availability of sequence information for other species, another aim of the project is to develop statistical methods to unravel the genetic secret of genomic imprinting in sequence level. Efficient statistical algorithms, biologically meaningful hypothesis tests, and robust model assessment tools are under investigation.
Imprinting phenomena have been increasingly observed in a wide spectrum, spanning from plants, animals to humans. In cereals, the endosperm of a grain is the main storage organ serving the major source of food for humans. A number of endosperm traits beneficial to humans are controlled by imprinted genes. In humans, many previously puzzling diseases such as Prader-Willis syndrome and Angelman syndrome are known to be affected by imprinted genes. In this project, the investigator will develop biologically meaningful statistical methods to hunt for imprinted genes underlying complex traits, in order to enhance our understanding of genetic architecture of genomic imprinting and the function of imprinted genes. The developed models, algorithms and software will allow researchers to better analyze their data in the hope of better understanding the genetic basis of genomic imprinting. The project will significantly benefit society by advancing the discovery of imprinted genes to help animal and plant breeders to improve trait quality, and to facilitate identification of new drugs to enhance public health. The research will be integrated into education to train new generations in statistical genetics, and will be widely disseminated through publications, presentations, online software and collaborations with geneticists.
