Weighted Epistatic Analysis of NSAIDs Hypersensitivity Data. (June 2017)

• Record Type: Journal Article
• Authors: Upton, Alex; Blanca, Miguel; Cornejo-García, José Antonio; Perkins, James Richard
• Is Part Of: Engineering applications of artificial intelligence. Volume 62(2017:Feb.)
• Page Start: 312
• Page End: 319
• Publication Date: 2017-06
• Subjects: NSAIDs hypersensitivity -- Epistasis -- GWAS -- Allergy -- HPC; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2016.09.007
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: Complex diseases such as allergy are thought to partly result from combinations of particular genetic variants, as well as additive effects of single variations acting independently. As a result, employing an epistatic interaction approach that focuses on identifying multiple single nucleotide polymorphism (SNP) interactions can build on genome wide association studies that focus on discovering associations between disease and individual variants, and can provide insights about the underlying disease mechanisms. In previous work, we identified a number of SNPs and genes potentially involved in nonsteroidal anti-inflammatory drugs (NSAIDs) hypersensitivity through the application of an epistatic analysis approach. In this study, we build on these approaches and use a weighted approach to identify additional SNPs and genes associated with this disorder. This is achieved through the implementation of a novel two stage weighted epistatic analysis approach. In the first step, epistatic analysis is carried out to identify SNP pairs associated with NSAIDs hypersensitivity, and weighted SNP interaction networks inferred based on their p-value. In the second step, these SNPs are mapped to their closest protein coding gene within a 500 Kb flanking distance, with a penalty applied to interactions involving SNPs not located within a gene, and gene interaction networks are constructed from this data. These networks are analysed using graph theory metrics, leading to the identification of several combinations of SNPs and genes potentially involved in and predictive of NSAIDs hypersensitivity. A number of potential asthma and atopy related genes are identified, such as KCNB2, as well as the gene CGNL1, which is differentially expressed following aspirin intake. In addition, subsequent pathway analysis of the gene interaction subnets uncovers significant enrichment for a number of biological pathways with a potential role in NSAIDs hypersensitivity, such as ALK1 and TGF-beta signalling, both associated with allergy. This study shows that applying a weighted epistatic analysis approach can provide further insights into the underlying mechanisms of NSAIDs hypersensitivity.