Multifactor Dimensionality Reduction Methodology and Related Software Package for Detecting andCharacterizing Gene-Gene and/or Gene-Environment Interactions- A Review

سال انتشار: 1394
نوع سند: مقاله کنفرانسی
زبان: انگلیسی
مشاهده: 515

نسخه کامل این مقاله ارائه نشده است و در دسترس نمی باشد

استخراج به نرم افزارهای پژوهشی:

لینک ثابت به این مقاله:

شناسه ملی سند علمی:

NASTARANCANSER01_082

تاریخ نمایه سازی: 26 شهریور 1395

چکیده مقاله:

The effects of single-nucleotide polymorphisms (SNPs) on a wide range ofdiseases/phenotypes with variable results have extensively been analyzed, however, in mostcases with a large proportion of genetic factors, these effects remained unexplained. It isdemonstrated that these limitations are due to the analytical strategy which limits analyses toonly individual SNPs, and thus it is becoming more feasible to evaluate the challenge ofdetecting SNP-SNP interactions. Detecting interactions among variables is a well knownproblem in data mining and human genetics. In addition, by focusing on associationsbetween SNPs and phenotypes, genome-wide association studies (GWAS) have beensuccessfully directed to identifying disease susceptibility genes for complex human diseases.Furthermore, one of the most challenges in GWAS is to detect gene-gene andgene†environment interactions. In this respect, Ritchie et al. (2001) proposed anonparametric and model-free genetic assessment called multifactor dimensionalityreduction (MDR). It was the ï rst machine learning approach speciï  cally designed foridentifying, characterizing and interpreting non-additive gene†gene interactions when thereis lack of statistically signiï cant independent major effects. Moreover, it provides anopportunity to classify multi-locus genotypes into high-risk and low-risk groups. The programcan be applied to analyze interactions among 2†15 genetic and/or environmental riskfactors especially in multifactorial disorders such as sporadic breast cancer. The main goal ofMDR is to change the presentation of the data using a useful induction algorithm to make nonadditiveinteractions easier to be analysed with other statistical methods such as logisticregression.

نویسندگان

Amir Tajbakhsh

Mashhad University of Medical Sciences, Mashhad, Iran

Fahimeh Afzal Javan

Mashhad University of Medical Sciences, Mashhad, Iran

Mahdi Rivandi

Mashhad University of Medical Sciences, Mashhad, Iran

Alireza Pasdar

Mashhad University of Medical Sciences, Mashhad, Iran