A biologically informed method for detecting associations with rare variants

Carrie C. Buchanan, John R. Wallace, Alex T. Frase, Eric S. Torstenson, Sarah A. Pendergrass, Marylyn D. Ritchie

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

With the recent flood of genome sequence data, there has been increasing interest in rare variants and methods to detect their association to disease. Many of these methods are collapsing strategies which bin rare variants based on allele frequency and functional predictions; but at this point, most have been limited to candidate gene studies with a small number of candidate genes. We propose a novel method to collapse rare variants based on incorporating biological information from the public domain. This paper introduces the functionality of BioBin, a biologically informed method to collapse rare variants and detect associations with a particular phenotype. We tested BioBin using low coverage data from the 1000 Genomes Project and discovered appropriate binning characteristics based on what one might expect given the size of the gene. We also tested BioBin using the pilot targeted exome data from 1000 Genomes Project. We used biologically-informed binning and differences in minor allele frequencies as a means to distinguish between two ancestral populations. Although BioBin is still in developmental stages, it will be a useful tool in analyzing sequence data and uncovering novel associations with complex disease.

Original languageEnglish (US)
Title of host publicationEvolutionary Computation, Machine Learning and Data Mining in Bioinformatics - 10th European Conference, EvoBIO 2012, Proceedings
Pages201-210
Number of pages10
DOIs
StatePublished - 2012
Event10th European Conference on Evolutionary Computation, Machine Learning and Data Mining in Bioinformatics, EvoBIO 2012 - Malaga, Spain
Duration: Apr 11 2012Apr 13 2012

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7246 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other10th European Conference on Evolutionary Computation, Machine Learning and Data Mining in Bioinformatics, EvoBIO 2012
Country/TerritorySpain
CityMalaga
Period4/11/124/13/12

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'A biologically informed method for detecting associations with rare variants'. Together they form a unique fingerprint.

Cite this