Calendar Name: seminars
Scheduled for: Monday, October 3 2011, 11:00 - 12:30
Event text: Dr Arthur Korte
Gregor Mendel Institute of Molecular Plant Biology
Vienna
AUSTRIA
Details: �Combining multiple traits for Genome-wide association
mapping�
ABSTRACT
Genome-wide association studies (GWAS) have become an
obvious general approach for studying the genetics of
natural variation. Although a plethora of significant
associations of SNPs with different traits have been
recently described (some of which have already been
verified experimentally), these analyses have been
focused on the identification of associations within a
single trait. When inbred lines are available, it is
possible to study many different traits for the same
genotype under different experimentally conditions. An
obvious next step is to derive a GWAS model that enables
the joint analysis of different traits, while accounting
for population structure. Hence we derived an extended
version of the broadly used mixed model and validated
our statistical model through simulations. We could
verify that our model both increase the power to
identify new SNPs and control the false discovery rate.
Due to the simultaneous testing of both the gene and the
gene-environment effect in the model we can identify
both, SNPs that are globally important in both traits
(not influenced by the environment) and SNPs important
for gene-environment interaction (having different
effects in the respective trait(s)). This method opens
the possibility to simultaneously analyze related traits
and gain insights into their underlying genetic
architecture. Applying the model on real data leads to
the identification of novel associations, which were not
identified by the marginal analysis
