Principles for the post-GWAS functional characterisation of cancer risk loci Matthew L. Freedman 1 , Alvaro N. A. Monteiro 2 , Simon A Gayther 3 , Gerhard A. Coetzee 4 , Angela Risch 5 , Christoph Plass 5 , Graham Casey 6 , Mariella De Biasi 7 , Chris Carlson 8 , Dave Duggan 9 , Michael James 10 , Pengyuan Liu 10 , Jay W. Tichelaar 10 , Haris G.Vikis 10 , Ming You 10 , Ian G.Mills 11 *

Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts, USA.
Nature Genetics (Impact Factor: 29.35). 06/2011; 43(6):513-8. DOI: 10.1038/ng.840
Source: PubMed
Download full-text


Available from: Ian G Mills, Oct 07, 2015
84 Reads
  • Source
    • "Chromatin association methods have been used to propose the identity of target genes (Freedman et al. 2011), while the selection of putative causal variants for detailed functional characterization has relied on fine mapping and colocalization of genetic variants with markers of active chromatin and/or binding sites for specific transcription factors. In particular, Cowper-Sal lari et al. (2012) demonstrated that breast cancer risk SNPs and their associated variant sets (AVS) preferentially localized to regions of H3K4me1 modification, FOXA1, and ESR1 binding. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Genome-wide association studies have identified over 70 common variants that are associated with breast cancer risk. Most of these variants map to non-protein-coding regions and several map to gene deserts, regions of several hundred kb lacking protein-coding genes. We hypothesised that gene deserts harbour long range regulatory elements that can physically interact with target genes to influence their expression. To test this, we developed Capture Hi-C (CHi-C) which, by incorporating a sequence capture step into a Hi-C protocol, allows high resolution analysis of targeted regions of the genome. We used CHi-C to investigate long range interactions at three breast cancer gene deserts mapping to 2q35, 8q24.21 and 9q31.2. We identified interaction peaks between putative regulatory elements ('bait fragments') within the captured regions and 'targets' that included both protein coding genes and long non-coding (lnc)RNAs, over distances of 6.6kb to 2.6Mb. Target protein-coding genes were IGFBP5, KLF4, NSMCE2 and MYC and target lncRNAs included DIRC3, PVT1 and CCDC26. For one gene desert we were able to define two SNPs (rs12613955 and rs4442975) that were highly correlated with the published risk variant and that mapped within the bait end of an interaction peak. In vivo ChIP-qPCR data show that one of these, rs4442975, affects the binding of FOXA1 and implicate this SNP as a putative functional variant.
    Genome Research 08/2014; 24(11). DOI:10.1101/gr.175034.114 · 14.63 Impact Factor
  • Source
    • "Most disease-associated alleles contribute to disease risk by acting as expression quantitative trait loci (eQTLs), influencing the expression or stability of a transcript [6-8]. In OA an excellent example of this is rs143383, which is located in the 5′ untranslated region (UTR) of GDF5; the T allele of this SNP correlates with reduced GDF5 expression in the joint tissues of patients with OA [9]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background An osteoarthritis (OA) susceptibility locus has been mapped to chromosome 3p21, to a region of high linkage disequilibrium encompassing twelve genes. Six of these genes are expressed in joint tissues and we therefore assessed whether any of the six were subject to cis-acting regulatory polymorphisms active in these tissues and which could therefore account for the association signal. Methods We measured allelic expression using pyrosequencing assays that can distinguish mRNA output from each allele of a transcript single nucleotide polymorphism. We assessed RNA extracted from the cartilage and other joint tissues of OA patients who had undergone elective joint replacement surgery. A two-tailed Mann–Whitney exact test was used to test the significance of any allelic differences. Results GNL3 and SPCS1 demonstrated significant allelic expression imbalance (AEI) in OA cartilage (GNL3, mean AEI = 1.04, p = 0.0002; SPCS1, mean AEI = 1.07, p < 0.0001). Similar results were observed in other tissues. Expression of the OA-associated allele was lower than that of the non-associated allele for both genes. Conclusions cis-acting regulatory polymorphisms acting on GNL3 and SPCS1 contribute to the OA association signal at chromosome 3p21, and these genes therefore merit further investigation.
    BMC Medical Genetics 05/2014; 15(1):53. DOI:10.1186/1471-2350-15-53 · 2.08 Impact Factor
  • Source
    • "In light of genome-wide association studies (GWAS), meQTL facilitate the interpretation of non-coding genetic variability and their association to phenotypic differences (Freedman et al., 2011; Hernandez and Singleton, 2012; Kilpinen and Dermitzakis, 2012). Recent studies have given an outlook of the potential of integrative genome-epigenome studies for the meaningful interpretation of genetic risk alleles (Gamazon et al., 2013; Scherf et al., 2013). "
    [Show abstract] [Hide abstract]
    ABSTRACT: With rapid advances in sequencing technologies, we are undergoing a paradigm shift from hypothesis- to data-driven research. Genome-wide profiling efforts have given informative insights into biological processes; however, considering the wealth of variation, the major challenge still remains in their meaningful interpretation. In particular sequence variation in non-coding contexts is often challenging to interpret. Here, data integration approaches for the identification of functional genetic variability represent a possible solution. Exemplary, functional linkage analysis integrating genotype and expression data determined regulatory quantitative trait loci and proposed causal relationships. In addition to gene expression, epigenetic regulation and specifically DNA methylation was established as highly valuable surrogate mark for functional variance of the genetic code. Epigenetic modification has served as powerful mediator trait to elucidate mechanisms forming phenotypes in health and disease. Particularly, integrative studies of genetic and DNA methylation data have been able to guide interpretation strategies of risk genotypes, but also proved their value for physiological traits, such as natural human variation and aging. This Review seeks to illustrate the power of data integration in the genomic era exemplified by DNA methylation quantitative trait loci. However, the model is further extendable to virtually all traceable molecular traits.
    Frontiers in Genetics 05/2014; 5:113. DOI:10.3389/fgene.2014.00113
Show more