TY - JOUR
T1 - A Multilocus Model of the Genetic Architecture of Autoimmune Thyroid Disorder, with Clinical Implications
AU - Vieland, Veronica J.
AU - Huang, Yungui
AU - Bartlett, Christopher
AU - Davies, Terry F.
AU - Tomer, Yaron
N1 - Funding Information:
This work was supported by National Institutes of Health grants DK61659, DK067555, and DK073681 to Y.T., DK052464 and DK069713 to T.F.D., and by the Research & Development Core of the Battelle Center for Mathematical Medicine at the Research Institute at Nationwide Children's Hospital.
PY - 2008/6/6
Y1 - 2008/6/6
N2 - We report here a preliminary model of the genetic architecture of Autoimmune Thyroid Disorder (AITD). Using a flexible class of mathematical modeling techniques, applied to an established set of data and supplemented with information both from candidate-gene and genome-wide-association studies and from basic bioinformatics, we find strong statistical support for a model in which AITD is the result of "hits" along three distinct genetic pathways: affected individuals have (1) a genetic susceptibility to clinical AITD, along with (2) a separate predisposition to develop the autoantibodies characteristic of AITD, and they also have (3) a predisposition to develop high levels of autoantibodies once they occur. Genes underlying each of these factors then appear to interact with one another to cause clinical AITD. We also find that a genetic variant in CTLA4 that increases risk for AITD in some people might actually protect against AITD in others, depending on which additional risk variants an individual carries. Our data show that the use of statistical methods for the incorporation of information from multiple sources, combined with careful modeling of distinct intermediate phenotypes, can provide insights into the genetic architecture of complex diseases. This model has several clinical implications, which we believe will prove relevant to other complex diseases as well.
AB - We report here a preliminary model of the genetic architecture of Autoimmune Thyroid Disorder (AITD). Using a flexible class of mathematical modeling techniques, applied to an established set of data and supplemented with information both from candidate-gene and genome-wide-association studies and from basic bioinformatics, we find strong statistical support for a model in which AITD is the result of "hits" along three distinct genetic pathways: affected individuals have (1) a genetic susceptibility to clinical AITD, along with (2) a separate predisposition to develop the autoantibodies characteristic of AITD, and they also have (3) a predisposition to develop high levels of autoantibodies once they occur. Genes underlying each of these factors then appear to interact with one another to cause clinical AITD. We also find that a genetic variant in CTLA4 that increases risk for AITD in some people might actually protect against AITD in others, depending on which additional risk variants an individual carries. Our data show that the use of statistical methods for the incorporation of information from multiple sources, combined with careful modeling of distinct intermediate phenotypes, can provide insights into the genetic architecture of complex diseases. This model has several clinical implications, which we believe will prove relevant to other complex diseases as well.
UR - http://www.scopus.com/inward/record.url?scp=44449155771&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=44449155771&partnerID=8YFLogxK
U2 - 10.1016/j.ajhg.2008.04.017
DO - 10.1016/j.ajhg.2008.04.017
M3 - Article
C2 - 18485327
AN - SCOPUS:44449155771
SN - 0002-9297
VL - 82
SP - 1349
EP - 1356
JO - American Journal of Human Genetics
JF - American Journal of Human Genetics
IS - 6
ER -