Genome-wide efficient attribute selection for purely epistatic models via Shannon entropy
by Amirhossein Manzourolajdad, Mohammad Saraee, Aghafakhr Mirlohi, Abolfazl Javan
International Journal of Business Intelligence and Data Mining (IJBIDM), Vol. 3, No. 4, 2008

Abstract: Epistasis plays an important role in the genetic architecture of common human diseases. Most complex diseases are believed to have multiple contributing loci that often have subtle patterns which make them fairly difficult to find in large data sets. Disorders that follow purely epistatic models cannot be detected by cases/control studies based on individual analysis of susceptible loci. The computational complexity of performing exhaustive searches for detecting such models in genome-wide applications is practically unfeasible. Furthermore, with ever-increasing number of both genotypes and individuals on one side, and little knowledge of complex traits on the other, it is becoming fairly difficult and time consuming to perform systematic genome-wide studies on such traits. We present and discuss a convenient framework for modelling epistasis using information theoretic concepts and algorithms inspired by such an approach. These generalised algorithms, which are especially in favour of purely epistatic models, are applied to both simulated and real data. The real data represents the genotype-phenotype values for Age-Related Macular Degeneration (AMD) disease. Many two-locus purely epistatic patterns were found for AMD. A new visualisation approach is also presented for the purpose of better illustrating epistasy for cases where the number of loci is more than two or three.

Online publication date: Sun, 25-Jan-2009

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Business Intelligence and Data Mining (IJBIDM):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com