Online Program Home
My Program

Abstract Details

Activity Number: 679 - Variable Selection and Prediction Models for Genomic Data
Type: Contributed
Date/Time: Thursday, August 2, 2018 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistics in Genomics and Genetics
Abstract #328499
Title: MDR with P Risk Scores Per Person with Application to Alzheimer's Disease Data
Author(s): Ye Li* and Richard Charnigo
Companies: University of Kentucky and University of Kentucky
Keywords: Multifactor Dimensionality Reduction; Risk Scores; Alzheimer's Disease ; gene-gene interaction effect
Abstract:

To identify genetic interactions for large dimensional data, multifactor dimensionality reduction (MDR) was developed by Ritchie et al in 2001, which entails characterizing each subject either as high or low risk. Extensions of MDR, such as quantitative MDR (Gui et al 2013), aggregated MDR (Dai et al 2013), and aggregated quantitative MDR (Crouch 2016), have been studied. Our work considers a situation where multiple interactions of a particular order (e.g., two way) may be considered simultaneously to obtain P (>1) risk scores to predict the continuous outcome for each subject.

Two methodologies are given in our work to search for a set of P risk scores to predict each subject's outcome when P is specified a priori and when P is not specified a priori, respectively. Using significant 2-way and 3-way SNP-SNP interactions found by AQMDR an QMDR(Crouch 2016), the first method is applied to ADNI data to select a set of 2 risk scores to predict CSF tau measurement for mild Alzheimer's disease status and to select a set of 3 risk scores to predict the measurement of CSF Abeta level for mild cognitive impairment status.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program