Page 93 - ASHG 2012 Annual Meeting Program Guide

Taking photographs or sound recordings in all meeting rooms is strictly prohibited. Thank you for your cooperation.
Thursday, November 8
Concurrent Invited Session II (21-28)
SESSION 21 – Mendelian Randomization: Using
Genetic Variants to Inform Causality in Observational
Room 135, Lower Level North, Moscone Center
David M. Evans, Univ. of Bristol, U.K.; Lyle
J. Palmer, Univ. of Ontario, Canada
A central problem in classical observational
epidemiology concerns whether an association
between a medically relevant exposure and a disease
outcome reflects a causal relationship, or whether the
purported association is a spurious result of latent
confounding, bias or reverse causality. For example,
does an inverse correlation between red wine drinking
and heart disease, mean that red wine prevents
heart attacks? The gold standard in proving whether
an association reflects a causal relationship is the
randomized controlled trial (RCT), in which participants
are randomly assigned into groups that differ on
average only in terms of their treatment (exposure)
effect. However, RCTs are typically expensive, of long
duration and in some cases, ethically or practically
infeasible to implement. Mendelian randomization
is a relatively new methodology that uses genetic
variants that have been robustly associated with
medically relevant exposures from genetic association
studies (e.g. genetic variants related to BMI, smoking,
alcohol etc.) to determine whether these exposures
are truly causally related to disease outcomes. This
session describes the principles behind Mendelian
randomization, explains how Mendelian randomization
can be used in epidemiological and gene expression
studies to assess causality, describes its assumptions
and limitations, and finally discusses extensions and
promising alternatives to the method.
Mendelian randomization: Overcoming
the limitations.
G . D. Smith. Univ. of Bristol, U.K.
Utilizing multiple variants to improve
Mendelian randomization studies.
B . Pierce. Univ.
of Chicago.
Application of Mendelian randomization
analyses in prospective studies from Denmark.
A. Tybjærg-Hansen. Univ. of Copenhagen, Denmark.
Mendelian randomization for HDL levels
and implications for clinical risk prediction.
B. F. Voight. Univ. of Pennsylvania Sch. of Med.
Thursday, November 8
Concurrent Invited Session II (21-28)
SESSION 22 – Common and Rare CNVs: Genesis,
Patterns of Variations and Human Diseases
Hall D, Lower Level North, Moscone Center
Moderators: Chack Yung Yu, Nationwide Children’s
Hosp. and The Ohio State Univ.; Edward J. Hollox,
Univ. of Leicester, U.K.
Copy number variation (CNV) is a part of human
genomic variation. There are two types of CNVs. The
first type is common CNVs that occur at regions with
complex genomic structures and engage genes
involving in gene-environment interactions. The
diversity of common CNVs both in terms of copy
number and sequence variation is remarkable, yet
because of their complex and challenging nature, they
are under-studied. The second type is de novo or
recent CNVs involved deletions and duplications of
genomic DNA fragments that are rare individually but
moderately frequent collectively, particularly among
subjects with developmental, neurologic and
psychiatric disorders. This session will highlight the
complexity of common CNVs and discuss their roles in
immune-mediated and cognitive diseases. In addition,
the session will feature mechanistic studies to
understand the generation of CNVs, and illustrate
techniques for the detection and quantification of
common and rare CNVs.
CNVs engaged in immune complex
handling and autoimmune diseases: Complement
C4 and immunoglobulin Fc-gamma receptors.
C. Y. Yu. Nationwide Children’s Hosp. and The Ohio
State Univ.
Human lineage-specific CNVs:
DUF 1220
domain copy number linked to cognitive disease
and brain evolution.
J. M. Sikela. Univ. of Colorado
Sch. of Med., Aurora.
TAR: A mixed genomic disorder caused
by a low-frequency regulatory SNP combined with
a 1q21.1 microdeletion.
W. H. Ouwehand. Cambridge
Univ. and Wellcome Trust Sanger Inst., Cambridge,
Genetic and environmental risk factors
for de novo CNVs.
T. W. Glover. Univ. of Michigan.
Frequency estimation of low-level
somatic mosaicism for pathogenic CNVs.
P. T. Stankiewicz. Baylor Col. of Med.