Characterization of Large Copy Number Variation in Mexican Type 2 Diabetes subjects

Abstract

The effect of Copy Number Variants (CNVs) on Type 2 Diabetes (T2D) remains little explored. The present study characterized large rare CNVs in 686 T2D and 194 non-T2D subjects of Mexican ancestry genotyped using the Affymetrix Genome-Wide Human SNP array 5.0. Rare CNVs with ≥ 100 kb length were identified using a stringent strategy based on merging CNVs calls generated using Birdsuit, iPattern and PennCNV algorithms. We applied three different strategies to evaluate the distribution of CNVs in the T2D and non-T2D samples: 1) Burden analysis, 2) Identification of CNVs in loci previously associated to T2D, and 3) Identification of CNVs observed only in the T2D group. In the CNV burden analysis, the T2D group showed a higher proportion of CNVs, and also a higher proportion of CNVs overlapping at least one gene than the non T2D group. Five of the six loci previously associated with T2D had duplications or deletions in the T2D sample, but not the non-T2D sample. A gene-set analysis including genes with CNVs observed only in the T2D group highlighted gene-sets related with sensory perception (olfactory receptors, OR) and phenylpyruvate tautomerase/dopachrome isomerase activity (MIF and DDT genes).

Figure 1

Analysis flowchart of the study’s aims.

Figure 2

Results of the gene-set analysis. The map shows a network of gene sets (nodes) related by mutual overlap (edges). A total of 130 genes derived from the genic regional analysis, the analysis of CNVs in candidate genes and the analysis of genic CNVs seen in the T2D group but not in the control group were used as input for the gene-set analysis, which identified 27 gene-sets. Blue circles corresponded to gene-sets with duplications and yellow circles represented gene-sets with duplications and deletions.