Abstract:

In complex diseases, such as coronary artery disease, the relationship of
genotype to phenotype is not one-to-one, but rather can vary as a function
of interactions with other genes and environments.  Even when a known gene
affects the phenotypic variation of interest, it explains only a small
portion of the total phenotypic variation.  Accordingly, in searching for
associations between genetic variation at candidate loci for coronary artery
disease, it is critical to augment the signal over the noise as much as
possible.  One method for increasing the genetic signal to noise ratio is to
use evolutionary history.  Mutations at a candidate gene occur in the
context of prior mutations.  At its time of creation, a new mutation will
exist on a specific haplotype background containing some, but not all, of
the mutational variants that had occurred earlier.  If recombination is rare
or is concentrated into hotspots, large segments of the candidate gene
sequence will accumulate mutational change in a manner that reflects their
evolutionary origin.  Any mutation that causes a phenotypic change will
therefore be imbedded in this evolutionary historical structure.  It will be
shown how this evolutionary history can be used to increase the power to
detect associations between DNA variants and phenotypes.   Two basic methods
will be outlined: nested clade analysis and tree scanning.