Analytic and Computational Challenges in the Identification of the Genetic Variation
Contributing to Complex Genetic Disorders

May 22, 2000

Nancy Cox
Department of Human Genetics
The University of Chicago


ABSTRACT

    The paradigm used to identify genes for traits with simple, Mendelian patterns
of transmission will clearly have to be modified if we are to be successful in
identifying genetic variation that affects susceptibility to common disorders with
complex patterns of transmission.  Such disorders are thought to arise from the
actions and interactions of many genetic and non-genetic factors.  The initial
stage of identification of the contributing genetic variation is its localization
through linkage and/or linkage disequilibrium mapping approaches.  To date, the
results of such studies in complex disorders would have to be characterized as
disappointing.  A key problem may be that even the most robust of the methods
currently used for localization ignore the gene x gene and gene x environment
interactions that, by definition, characterize the familial component of such
disorders.  We have developed a number of approaches for considering the effects
of gene x gene interaction in the context of genome-wide screens for localization
of genes influencing susceptibility for complex traits, and I will present some of
our most recent results.  Once regions likely to contain suceptibility genes for
complex traits have been successfully identified, the major analytic problem is
in identifying the genetic variation actually affecting susceptibility to disease
and distinguishing that variation from the nearby genetic variation that will be
in linkage disequilibrium.  In studies focussed on the positional cloning of a
gene for type 2 diabetes, we developed a number of analytic strategies designed
to consider not only the association of the genetic variation with disease, but
also the association of genetic variation with the original evidence for linkage,
which may improve the ability to distinguish the causal variation.  However, the
genetic model suggested by our results to date is complex, including elements of
allelic epistasis.  These results, in turn, suggest additional analytic and
computational challenges for the identification of genetic variation contributing
to complex genetic disorders.