Send to:

Choose Destination
See comment in PubMed Commons below
Hum Hered. 2010;69(4):242-53. doi: 10.1159/000298285. Epub 2010 Mar 26.

Accounting for disease model uncertainty in mapping heterogeneous traits--a Bayesian model averaging approach.

Author information

  • 1Department of Biostatistics, School of Public Health, University of North Texas Health Science Center, Fort Worth, TX, USA.



Locus heterogeneity, wherein a disease can be caused in different individuals by different genes and/or environmental factors, is a ubiquitous feature of complex traits. A Bayesian approach has been proposed to account for variable rates of heterogeneity across families in a parametric linkage analysis setup [Biswas and Lin: J Am Stat Assoc 2006;101:1341-1351]. As with any parametric approach, its application requires specification of the disease model, which limits its practical utility.


We address this limitation by proposing a Bayesian model averaging (BMA) approach. We consider a finite number of disease models and treat the model as an unknown parameter. In practice, we use simple single-locus disease models as various categories for model.


Our simulations as well as analysis of Genetic Analysis Workshop 13 simulated data show that BMA retains at least 80% of the power that is obtained by analyzing under the true disease model. The coverage probability of interval for disease gene is maintained around the nominal level. Finally, we apply BMA to a Late-Onset Alzheimer's Disease dataset and find evidence for linkage on chromosomes 19, 9, and 21.


We conclude that the BMA approach utilizing simple single-locus models for averaging is effective for mapping heterogeneous traits.

Copyright 2010 S. Karger AG, Basel.

[PubMed - indexed for MEDLINE]
Free PMC Article
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for S. Karger AG, Basel, Switzerland Icon for PubMed Central
    Loading ...
    Write to the Help Desk