Format

Send to

Choose Destination
Hum Mol Genet. 2016 Oct 1;25(19):4339-4349. doi: 10.1093/hmg/ddw283. Epub 2016 Aug 24.

Collapsed methylation quantitative trait loci analysis for low frequency and rare variants.

Author information

1
MRC Integrative Epidemiology Unit (IEU), School of Social and Community Medicine, University of Bristol, Oakfield House, Oakfield Grove, Bristol, UK.
2
University of Exeter Medical School, University of Exeter, Exeter, UK.
3
Institute of Psychiatry, King's College London, London, UK.
4
Department of Twin Research and Genetic Epidemiology, King's College London, London, UK.
5
Avon Longitudinal Study of Parents and Children (ALSPAC) & School of Social and Community Medicine, University of Bristol, Bristol, UK.
6
Bristol Genetic Epidemiology Laboratories, School of Social and Community Medicine, University of Bristol, Bristol, UK.
7
Intelligent Systems Laboratory, University of Bristol, Bristol, UK.
8
MRC Integrative Epidemiology Unit (IEU), School of Social and Community Medicine, University of Bristol, Oakfield House, Oakfield Grove, Bristol, UK Tom.Gaunt@bristol.ac.uk.

Abstract

BACKGROUND:

Single variant approaches have been successful in identifying DNA methylation quantitative trait loci (mQTL), although as with complex traits they lack the statistical power to identify the effects from rare genetic variants. We have undertaken extensive analyses to identify regions of low frequency and rare variants that are associated with DNA methylation levels.

METHODS:

We used repeated measurements of DNA methylation from five different life stages in human blood, taken from the Avon Longitudinal Study of Parents and Children (ALSPAC) cohort. Variants were collapsed across CpG islands and their flanking regions to identify variants collectively associated with methylation, where no single variant was individually responsible for the observed signal. All analyses were undertaken using the sequence kernel association test.

RESULTS:

For loci where no individual variant mQTL was observed based on a single variant analysis, we identified 95 unique regions where the combined effect of low frequency variants (MAF ≤ 5%) provided strong evidence of association with methylation. For loci where there was previous evidence of an individual variant mQTL, a further 3 regions provided evidence of association between multiple low frequency variants and methylation levels. Effects were observed consistently across 5 different time points in the lifecourse and evidence of replication in the TwinsUK and Exeter cohorts was also identified.

CONCLUSION:

We have demonstrated the potential of this novel approach to mQTL analysis by analysing the combined effect of multiple low frequency or rare variants. Future studies should benefit from applying this approach as a complementary follow up to single variant analyses.

PMID:
27559110
PMCID:
PMC5291201
DOI:
10.1093/hmg/ddw283
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for Silverchair Information Systems Icon for PubMed Central
Loading ...
Support Center