Congruence as a measurement of extended haplotype structure across the genome

<p>Abstract</p> <p>Background</p> <p>Historically, extended haplotypes have been defined using only a few data points, such as alleles for several HLA genes in the MHC. High-density SNP data, and the increasing affordability of whole genome SNP typing, creates the opportunity to define higher resolution extended haplotypes. This drives the need for new tools that support quantification and visualization of extended haplotypes as defined by as many as 2000 SNPs. Confronted with high-density SNP data across the major histocompatibility complex (MHC) for 2,300 complete families, compiled by the Type 1 Diabetes Genetics Consortium (T1DGC), we developed software for studying extended haplotypes.</p> <p>Methods</p> <p>The software, called ExHap (Extended Haplotype), uses a similarity measurement we term congruence to identify and quantify long-range allele identity. Using ExHap, we analyzed congruence in both the T1DGC data and family-phased data from the International HapMap Project.</p> <p>Results</p> <p>Congruent chromosomes from the T1DGC data have between 96.5% and 99.9% allele identity over 1,818 SNPs spanning 2.64 megabases of the MHC (<it>HLA-DRB1 </it>to <it>HLA-A</it>). Thirty-three of 132 DQ-DR-B-A defined haplotype groups have > 50% congruent chromosomes in this region. For example, 92% of chromosomes within the DR3-B8-A1 haplotype are congruent from <it>HLA-DRB1 </it>to <it>HLA-A </it>(99.8% allele identity). We also applied ExHap to all 22 autosomes for both CEU and YRI cohorts from the International HapMap Project, identifying multiple candidate extended haplotypes.</p> <p>Conclusions</p> <p>Long-range congruence is not unique to the MHC region. Patterns of allele identity on phased chromosomes provide a simple, straightforward approach to visually and quantitatively inspect complex long-range structural patterns in the genome. Such patterns aid the biologist in appreciating genetic similarities and differences across cohorts, and can lead to hypothesis generation for subsequent studies.</p>

Abstract licence: CC BY-NC-ND

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.