Analysis of heterogeneity and epistasis in physiological mixed populations by combined structural equation modelling and latent class analysis.
Biological systems are interacting networks of genetic and molecular entities in which genetic variation generates phenotypic heterogeneity. Usually the heterogeneity is modelled as a dichotomous trait (e.g.
affected versus non-affected). This is far too simplistic considering the complexity and genetic variations of the networks.
In this study heterogeneity is resolved in a latent class framework combined with structural equation modelling using phenotypic indicators of distinct physiological processes. Here the metabolic syndrome, which is known to be a heterogenous, polygenic condition with a clinical endpoint (type 2 diabetes mellitus), is modelled.
In the model presented here genetic factors are not included and no genetic model is assumed except that genes operate in networks. The impact of stratification of the study population on genetic interaction is demonstrated by including several genes previously associated with the metabolic syndrome and type 2 diabetes mellitus.
Results: The analysis revealed 19 distinct subpopulations with different propensity for developing diabetes mellitus in a large healthy study population .
The allocation of subjects into subpopulations were very accurate with an entropy measure of nearly 0.9. With a few exceptions none of the genes included in the study showed association to the metabolic syndrome.
However, when analysed for epistasis approximately one third of all possible interactions were highly significant except for two polymorphisms that did not interact with any other polymorphism. In particular, the number of interactions increased after stratifying the study population, suggesting that interactions are masked in heterogenous populations.
In addition, the genetic variance increased 35-fold on average in the subpopulations.
Conclusions: The major conclusions from this study are that 1) it is mandatory to define physiological homogenous subpopulations to determine the genetics of the metabolic process, and 2) that search for single-gene effects is generally of very low power; whereas epistasis (i.e. genetic networks) are ubiquitous and should be the basis of modelling any biological process.
Author: Mogens Fenger, Allan Linneberg, Thomas Werge and Torben Jorgensen
Credits/Source: BMC Genetics 2008, 9:43
Published on: 2008-07-08
Copyright by the authors listed above - made available via BioMedCentral (Open Access). Please
make sure to read our disclaimer prior to contacting 7thSpace Interactive. To contact our editors, visit our online helpdesk. If you wish submit your own press release, click here.
Social Bookmarking
Digg this! | Post to del.icio.us | Post to Furl | Add to Netscape | Add to Yahoo! | Rojo
|
|
