Identification of multiple gene-gene interactions for ordinal phenotypes

Min-Seok Kwon,Sohee Oh,Taesung Park,Kyunga Kim

doi:10.1186/1755-8794-6-s2-s9

Min-Seok Kwon, Sohee Oh + Show 2 more

Open Access

https://doi.org/10.1186/1755-8794-6-s2-s9

Copy DOI

Abstract

BackgroundMultifactor dimensionality reduction (MDR) is a powerful method for analysis of gene-gene interactions and has been successfully applied to many genetic studies of complex diseases. However, the main application of MDR has been limited to binary traits, while traits having ordinal features are commonly observed in many genetic studies (e.g., obesity classification - normal, pre-obese, mild obese and severe obese).MethodsWe propose ordinal MDR (OMDR) to facilitate gene-gene interaction analysis for ordinal traits. As an alternative to balanced accuracy, the use of tau-b, a common ordinal association measure, was suggested to evaluate interactions. Also, we generalized cross-validation consistency (GCVC) to identify multiple best interactions. GCVC can be practically useful for analyzing complex traits, especially in large-scale genetic studies.Results and conclusionsIn simulations, OMDR showed fairly good performance in terms of power, predictability and selection stability and outperformed MDR. For demonstration, we used a real data of body mass index (BMI) and scanned 1~4-way interactions of obesity ordinal and binary traits of BMI via OMDR and MDR, respectively. In real data analysis, more interactions were identified for ordinal trait than binary traits. On average, the commonly identified interactions showed higher predictability for ordinal trait than binary traits. The proposed OMDR and GCVC were implemented in a C/C++ program, executables of which are freely available for Linux, Windows and MacOS upon request for non-commercial research institutions.

Highlights

Multifactor dimensionality reduction (MDR) is a powerful method for analysis of gene-gene interactions and has been successfully applied to many genetic studies of complex diseases
We propose an ordinal MDR (OMDR) approach that enables one to analyze a joint effect of multiple genetic variables on an ordinal categorical trait
The overall best OMDR classifiers are selected based on the predictability and generalized CVC based on top-K selection (GCVCK) among the best ones for various values of m, which result from the previous steps

Summary

Introduction

Multifactor dimensionality reduction (MDR) is a powerful method for analysis of gene-gene interactions and has been successfully applied to many genetic studies of complex diseases. Many different methods have been proposed to analyze gene-gene interactions in genetic association studies [4,5], and can be categorized to methods based on regression modeling [6-9], pattern recognition [10,11], and data reduction [12-14]. Machine learning approaches, While each method has its own advantages and disadvantages, the multifactor dimensionality reduction (MDR) method, a data-reduction approach, is known to have the advantages in examining high-order interactions and detecting interactions without main effects [13,18-20], and has been widely applied to detect gene-gene interactions in many common diseases (see the related literature available on http://epistasis.org). The applicability of existing MDR approaches is still restricted mainly to binary traits

Methods

Results

Conclusion