Abstract:
|
Identifying the genetic basis of complex traits is an important problem with the potential to impact a broad range of biological endeavors. A number of good statistical methods are available for quantitative trait loci (QTL) mapping that allow for the efficient identification of multiple, potentially interacting, loci under a variety of experimental conditions. Although proven useful in hundreds of studies, the majority of these methods assume a single model common to each subject and consequently sacrifice power and accuracy when genetically distinct subclasses exist. To address this, we have developed an approach to enable latent class QTL mapping. The approach combines the idea of latent class regression with stepwise variable selection and traditional QTL mapping to estimate the number of subclasses in a population, and to identify the genetic model that best describes each subclass. Simulation demonstrated good performance of the method when latent classes are present and when they are not, with accurate estimation of QTLs. Application of the method to a study of diabetes in mouse gives insight into the genetic bases of diabetes-related complex traits.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.