An Optimum Random Forest Model for Prediction of Genetic Susceptibility to Complex Diseases

扫码查看

原文链接

NETL

外文摘要：High-throughput single nucleotide polymorphism (SNP) genotyping technologies make massive genotype data， with a large number of individuals， publicly available。 Accessibility of genetic data makes genome-wide association studies for complex diseases possible。 One of the most challenging issues in genome-wide association studies is to search and analyze genetic risk factors resulting from interactions of multiple genes。 The integrated risk factor usually have a higher risk rate than single SNPs。 This paper explores the possibility of applying random forest to search disease-associated factors for given case/control samples。 An optimum random forest based algorithm is proposed for the disease susceptibility prediction problem。 The proposed method has been applied to publicly available genotype data on Crohn's disease and autoimmune disorders for predicting susceptibility to these diseases。 The achieved accuracy of prediction is higher than those achieved by universal prediction methods such as Support Vector Machine (SVM) and previous known methods。

外文关键词：

random forestassociation studycomplex diseasessusceptibilityrisk factorprediction

作者：

Weidong Mao、Shannon Kelly

展开 >

作者单位：

Department of Computer Science, Shippensburg University, Shippensburg, PA 17257

会议名称：

Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining(PAKDD 2007)

会议地点：

Nanjing(CN)

会议母体文献：

Advances in Knowledge Discovery and Data Mining; Lecture Notes in Artificial Intelligence; 4426

页码：

193-204

出版时间：

2007