From SNPs to pathways: Biological interpretation of type 2 diabetes (T2DM) genome wide association study (GWAS) results

Elisa Cirillo*, Martina Kutmon, Manuel Gonzalez Hernandez, Tom Hooimeijer, Michiel E. Adriaens, Lars M. T. Eijssen, Laurence D Parnell, Susan L. Coort, Chris T. Evelo

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Genome-wide association studies (GWAS) have become a common method for discovery of gene-disease relationships, in particular for complex diseases like Type 2 Diabetes Mellitus (T2DM). The experience with GWAS analysis has revealed that the genetic risk for complex diseases involves cumulative, small effects of many genes and only some genes with a moderate effect. In order to explore the complexity of the relationships between T2DM genes and their potential function at the process level as effected by polymorphism effects, a secondary analysis of a GWAS meta-analysis is presented. Network analysis, pathway information and integration of different types of biological information such as eQTLs and gene-environment interactions are used to elucidate the biological context of the genetic variants and to perform an analysis based on data visualization. We selected a T2DM dataset from a GWAS meta-analysis, and extracted 1,971 SNPs associated with T2DM. We mapped 580 SNPs to 360 genes, and then selected 460 pathways containing these genes from the curated collection of WikiPathways. We then created and analyzed SNP-gene and SNP-gene-pathway network modules in Cytoscape. A focus on genes with robust connections to pathways permitted identification of many T2DM pertinent pathways. However, numerous genes lack literature evidence of association with T2DM. We also speculate on the genes in specific network structures obtained in the SNP-gene network, such as gene-SNP-gene modules. Finally, we selected genes relevant to T2DM from our SNP-gene-pathway network, using different sources that reveal gene-environment interactions and eQTLs. We confirmed functions relevant to T2DM for many genes and have identified some-LPL and APOB-that require further validation to clarify their involvement in T2DM.

Original languageEnglish
Article number0193515
Pages (from-to)e0193515
Number of pages19
JournalPLOS ONE
Volume13
Issue number4
DOIs
Publication statusPublished - 4 Apr 2018

Keywords

  • Journal Article
  • RISK LOCI
  • VARIANTS
  • DISEASE
  • GENES
  • ONTOLOGY
  • UNIFICATION
  • NETWORKS
  • IDENTIFICATION
  • COMPLEX TRAITS
  • TOOL

Cite this