|
Nengjun
Yi, Department of Biostatistics, Section on Statistical Genetics,
University of Alabama, Birmingham, AL
Bayesian model selection for
mapping interacting QTL
Abstract:
Many complex traits are determined by multiple genetic and
environmental influences. Gene-gene and gene-environment interactions
play an important role in the genetic control and evolution of complex
traits. Identification
of genome-wide interacting quantitative trait loci (QTL) has been a
daunting challenge, mainly due to huge model space. In this study, we
propose a Bayesian model selection approach to identifying interacting
QTL across the entire genome for complex traits in experimental
crosses. The proposed method is able to model fixed or random
covariates, and simultaneously detect gene-gene and gene-environment
interactions. Computationally efficient Markov chain Monte Carlo (MCMC)
algorithms are developed to sample from the posterior distribution.
Statistical properties of the proposed algorithms are explored. We
detail how to use prior knowledge to specify prior distributions for
all unknowns. Several strategies to reduce the model space are
incorporated into the proposed approach, allowing more rapid
identification and exploration of important interactions. We
illustrate the proposed method using real data sets.
Nengjun Yi,
is an Associate Professor in the Department of Biostatistics at UAB.
His primary research interests include Bayesian statistics, Markov
chain Monte Carlo (MCMC) algorithms and statistical genetics. He has
published papers on these topics in journals such as PNAS, Genetics
and Genetical Research. Recently, his research has concentrated on
developing Bayesian model selection methods for genome-wide
interacting QTL analysis.
Guimin Gao, Department
of Biostatistics, Section on Statistical Genetics, University of
Alabama, Birmingham, AL
Haplotyping Methods
for QTL Mapping in Pedigrees
Abstract:
Haplotyping in
pedigrees is an important step for the fine mapping of Quantitative
Trait Loci (QTL) or complex disease genes. To reconstruct haplotypes
efficiently for large pedigrees with large numbers of linked loci, we
develop two algorithms based on conditional probabilities and
likelihood computations. The first algorithm (the conditional
probability method) produces a single, approximately optimal haplotype
configuration. The other algorithm (the conditional enumeration
method) identifies a set of haplotype configurations with high
probabilities conditional on the observed genotype data for a
pedigree. The algorithms are tested on published and simulated data
sets. The new haplotyping methods are much faster and
provide more information than several existing stochastic and
rule-based methods. The accuracies of the new methods are equivalent
to or better than those of these existing methods. The new method can
be used for fine mapping with joint linkage disequilibrium and linkage
analysis, which improves the power and accuracy of QTL mapping.
Biosketch:
Dr. Guimin Gao
received his Ph.D in Biostatistics from Sun Yat-sen University of
Medical Sciences, China, and joined the section of statistical
genetics at University of Alabama at Birmingham as a research
assistant professor in August 2005. His research focuses on 1) fine
mapping of Quantitative Trait Loci (QTL) and complex disease gene in
pedigrees, via joint linkage disequilibrium (LD) and linkage analysis
with the variance component method; 2) developing fine mapping
software for different population structures. A major focus of his
research is on developing efficient haplotyping methods and
identity-by-descent (IBD) matrix calculation methods for complex
pedigrees with large sizes and large numbers of linked loci.
Mark Beasley, Department
of Biostatistics, Section on Statistical Genetics, University of
Alabama, Birmingham, AL
Joint tests for quantitative
trait loci in transmission disequilibrium testing
Abstract:
Beasley et al.
(2004) proposed joint tests that capitalize on the information
available in both regression-based and marginal-based tests. These
methods increase the power for detecting QTL effects in the context of
a backcross and an F2 intercross of inbred strains, especially in
one-tailed extreme threshold (e.g., case-only) designs. Tiwari et al.
(2005) aptly note that the trios available for use in Allison’s (1997)
TDTQ5 can be viewed as a combined analysis of two separate
backcrosses and an F2 intercross. They extended the approach of
Beasley et al. (2004) to human genetics research in the context of
transmission disequilibrium testing. Although their OLS-based
procedure was effective for detecting linkage for additive modes of
inheritance, it was less powerful for non-additive models. Their
logistic regression procedure was effective for detecting linkage
under any mode of inheritance but suffers from its complicated form,
which also limits its extension into multi-locus models. Our purpose
is to modify these OLS-based methods in order to model non-additive
modes of inheritance, which will make extensions to multi-locus models
and multivariate phenotypes more straightforward.
Biosketch:
Dr. Beasley
received his Ph.D. in Statistics from the Southern Illinois
University, Carbondale, IL. Dr. Beasley is an Associate Professor of
Biostatics and a distinguished member of the Section on Statistical
Genetics. Dr. Beasley has a strong research record in the area of
statistical methodology, focused in five major areas: (1)
methodological problems in statistical genetics; (2) nonparametric
statistics; (3) simulation studies; (4) the use of linear models; and
(5) longitudinal analysis. He has extensive experience in admixture
mapping.
David T. Redden,
Department of Biostatistics, University of Alabama, Birmingham, AL.
Regional Admixture Mapping and Structured Association Testing:
Conceptual Unification using a General Linear Model
Abstract:
Individual genetic admixture estimates,
determined both across the genome and at specific genomic regions,
have been proposed for use in identifying specific genomic regions
harboring loci influencing phenotypes in regional admixture mapping
(RAM). Estimates of individual ancestry can be used in structured
association tests (SAT) to reduce confounding induced by various forms
of population substructure. Although presented as two distinct
approaches, we provide a conceptual framework in which both RAM and
SAT are special cases of a more general linear model which allows for
greater modeling flexibility, adaptation to multiple designs,
inclusion of covariates and interaction terms, and multi-locus models.
We clarify which variables are sufficient upon which to condition in
order to prevent spurious associations. This approach should allow far
wider use of RAM and SAT, often using standard software, in addressing
admixture as either a confounder of association studies or a tool for
finding loci influencing complex phenotypes in species as diverse as
plants, humans, and non-human animals.
Biosketch:
Dr. Redden
received his Ph.D. in Applied Statistics from the University of
Alabama, Tuscaloosa. Currently, he is a faculty in the Department of
Biostatistics (Section on Statistical Genetics) at UAB. His research
interests include Genetic Association studies, estimation of
population admixture, multisatge designs, and clinical trials. His
primary academic interest is the effect of population heterogeneity
upon genetic association studies.
|
