Parallel and serial computing tools for testing single-locus and epistatic SNP effects of quantitative traits in genome-wide association studies

Ma, Li; Runesha, Hakizumwami Birali; Dvorkin, Daniel; Garbe, John; Da, Yang

doi:10.1186/1471-2105-9-315

Cited by 98 publications

(118 citation statements)

References 17 publications

(12 reference statements)

Supporting

Mentioning

118

Contrasting

Order By: Relevance

“…According to [9], it would require 1.2 years to do the pairwise epistasis testing of 500,000 SNPs using the serial program on a 2.66 GHz single processor without parallel processing. In this paper, we have provided a cloud epistatic computing model (CEO) for large scale epistatic interactions using the MapReduce framework.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

CEO a cloud epistasis computing model in GWAS

Wang

Tan

et al. 2010

2010 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

View full text Add to dashboard Cite

Abstract-The 1000 Genome project has made available a large number of single nucleotide polymorphisms (SNPs) for genome-wide association studies (GWAS). However, the large number of SNPs has also rendered the discovery of epistatic interactions of SNPs computationally expensive. Parallelizing the computation offers a promising solution. In this paper, we propose a cloud-based epistasis computing (CEO) model that examines all k-locus SNPs combinations to find statistically significant epistatic interactions efficiently. Our CEO model uses the MapReduce framework which can be executed both on user's own clusters or on a cloud environment. Our cloud-based solution offers elastic computing resources to users, and more importantly, makes our approach affordable and available to all end-users. We evaluate our CEO model on a cluster of more than 40 nodes. Our experiment results show that our CEO model is computationally flexible, scalable and practical.

show abstract

Section: Resultsmentioning

confidence: 99%

“…This is the approach we adopt in this paper. [9] provided a tool for processing only singlelocus and two-locus SNPs analysis using supercomputer system on selected types of processors and compilers. However, it is not easy for researchers to rewrite their own programs on specialized hardware.…”

Section: Introductionmentioning

confidence: 99%

CEO a cloud epistasis computing model in GWAS

Wang

Tan

et al. 2010

2010 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

View full text Add to dashboard Cite

show abstract

“…All these tests were investigated between mild pain versus moderate pain and mild pain versus severe pain (GLM procedure). SNP-SNP interactions in relation to sleep in musculoskeletal pain were observed from the genotype data implementing extended Kempthorne model by software epiSNP [13]. Single locus tests correspond to overall marker effect (M), dominant effect (D) and additive effect (A).…”

Section: Statistical Analysis:-mentioning

confidence: 99%

SNP-SNP interactions within catechol-O-methyltransferase (COMT) gene influence sleep quality in subjects having chronic musculoskeletal pain-A Genetic Exploration ofMusculoskeletal Pain Study (GEMPS).

Dhillon¹,

Khullar²,

Kaur³

et al. 2016

IJAR

View full text Add to dashboard Cite

“…Studies of model organisms suggested epistasis or gene-gene interactions to be a common phenomenon [13,14,15,16], and a number of gene-gene interactions have been reported in gene mapping studies in animals, plants, and other model organisms [17,18,19,20]. However, gene-gene interactions have proven difficult to find in humans [21,22], mainly due to low statistical power caused by the small effect size, the low minor genotype frequency of the multiple-SNP combinations, the large combinatorial number of interaction tests required [14,23], and the lack of control of environmental conditions. Hence, in order to improve the power of detection of gene-gene interactions in human GWAS, many approaches have been developed to prioritize candidate genes or SNPs using biological knowledge from established GWAS hits [6,24], protein-protein interactions (PPIs) [25,26], and pathway information [27].…”

Section: Introductionmentioning

confidence: 99%

Biological Knowledge-Driven Analysis of Epistasis in Human GWAS with Application to Lipid Traits

Keinan

Clark

2014

Methods in Molecular Biology

Self Cite

View full text Add to dashboard Cite

While the importance of epistasis is well established, specific gene-gene interactions have rarely been identified in human genome-wide association studies (GWAS), mainly due to the low power associated with such interaction tests. In this chapter, we integrate biological knowledge and human GWAS data to reveal epistatic interactions underlying quantitative lipid traits which are major risk factors for coronary artery disease. To increase power to detect interactions, we only tested pairs of SNPs filtered by prior biological knowledge, including GWAS results, protein-protein interactions, and pathway information. Using published GWAS results and 9,713 European Americans (EA) from the Atherosclerosis Risk in Communities (ARIC) study, we identified an interaction between HMGCR and LIPC affecting high-density lipoprotein cholesterol (HDL-C) levels. We then validated this interaction in additional EA cohorts from the Framingham Heart Study and the Multi-Ethnic Study of Atherosclerosis (MESA). We also validated the interaction in the ARIC African American sample and in the Hispanic American sample from MESA. Both HMGCR and LIPC are involved in the metabolism of lipid and lipoproteins, and LIPC itself has been associated with HDL-C. Moreover, the interaction affects HDL-C twice as much as the marginal effect of LIPC, with the effect of the two genes and their interaction combined explaining 0.8% of the variation in HDL. These results suggest the potential of the biological knowledge-driven approach to detect epistatic interactions in human GWAS, which may hold the key to exploring the role gene-gene interactions play in connecting genotype and phenotype in current and future genetic association studies.

show abstract

Parallel and serial computing tools for testing single-locus and epistatic SNP effects of quantitative traits in genome-wide association studies

Cited by 98 publications

References 17 publications

CEO a cloud epistasis computing model in GWAS

CEO a cloud epistasis computing model in GWAS

SNP-SNP interactions within catechol-O-methyltransferase (COMT) gene influence sleep quality in subjects having chronic musculoskeletal pain-A Genetic Exploration ofMusculoskeletal Pain Study (GEMPS).

Biological Knowledge-Driven Analysis of Epistasis in Human GWAS with Application to Lipid Traits

Contact Info

Product

Resources

About