The formatting of the blog posting may be odd if you are reading this in a Feed Reader or via e-mail distribution, so click on the title above (which is an active link) to view the website version.
FIGURE 1: Positions of Centromeres, Regions of Excess IBD, HLA Region, and SNP-Poor Regions (Build 36)
|
FIGURE 2: Positions of Centromeres, Regions of Excess IBD, HLA Region, and SNP-Poor Regions (Build 37)
|
I do a lot of chromosome mapping using Kitty Cooper's great Ancestor Mapping Tool. There are a few regions in the genome more likely to have issues when comparing matching DNA segments with others: at the centromeres and regions of excess IBD sharing (IBD = identical by descent). Instead of looking these up in tables, I've plotted them on chromosome maps, as visual representations are easier than looking up values in a table and they allow side-by-side comparisons to be made to my chromosome maps. I also added the Human Leukocyte Antigen (HLA) region on chromosome 6, which exhibits greater IBD sharing than expected, and the SNP-poor regions (SNP = single nucleotide polypeptide) rather than having blank areas on the chromosome map.
Figure 1 shows the positions for Build 36, which is the build used by Family Tree DNA (FTDNA) and GEDmatch. Figure 2 shows the positions for Build 37, the build used by 23andMe. The input files (.csv files) are available on the Downloads page, in the event you want to modify these, which include footnotes with additional notes.
Centromeres
For Build 36 (Figure 1), the centromere positions were taken from Family Tree DNA's FAQ Where are the centromeres located on each autosomal chromosome?, although the X chromosome centromere is missing. But thanks to a FTDNA Forum posting (Here), I found a source for the X chromosome centromere for Build 36 on one of the tables provided by USCS Genome Bioinformatics. The genetic lengths of the centromeres (in cM) for Build 36 were calculated using sex-averaged map positions from the Rutgers Map Interpolator (Laboratory of Computational Genetics, Rutgers University): these are generally very short (0 - 0.68 cM), but the genetic distance for the centromere on chromosome 22 is longer (5.31 cM).
For Build 37 (Figure 2), these were taken from the ISOGG Wiki page on the Centromere (data provided by Kitty Cooper), but I couldn't locate a map interpolator for Build 37 to calculate the centromere cMs for this build.
For Build 37 (Figure 2), these were taken from the ISOGG Wiki page on the Centromere (data provided by Kitty Cooper), but I couldn't locate a map interpolator for Build 37 to calculate the centromere cMs for this build.
Regions of Excess IBD
The ISOGG Wiki page Identical by Descent has a great summary of regions of excess IBD sharing, also referred to as pile-up regions. Table 1 (on the right) shows the 14 regions of the genome >5 cM with detected pairwise IBD at least 4-fold greater than expected using hg19 (Build 37) coordinates from Li et al [1], ordered by chromosome number – these are plotted in green in Figure 2. In addition, based on conversion using the LiftOver Tool from the UCSC, Build 36 (hg18) coordinates are also shown in Table 1 (on the left) and are plotted on Figure 1.
TABLE 1: Regions of Excess IBD for Build 36 and Build 37
HLA Region on Chromosome 6
|
The set of genes known as the Major Histocompatibility Complex (MHC), including the HLA genes, which are tested for organ transplant compatibility, are located between positions 29,750,000 to 33,100,000 on chromosome 6 [2]. The recombination rate is low in this area and exhibits greater IBD than expected. This region is often readily apparent on GEDmatch, as shown in the two examples on Figure 3. The position of the HLA region is plotted in blue on Figures 1 and 2.
|
FIGURE 3: Two examples of the HLA Region on Chromosome 6 (GEDmatch)
|
SNP-Poor Regions
|
The "SNP-poor" regions are areas where few SNPs are included on the chip and the largest of these are found at the beginning of chromosomes 13, 14, 15, 21, 22, and X. On FTDNA, these are grayed out and on GEDmatch the chromosome maps just exclude them. To prevent these areas from being blank when using Kitty Cooper's Ancestor Mapping Tool, I plot the SNP-poor regions on my chromosome maps (just like segments identified to specific ancestors), as shown on Figure 4.
|
FIGURE 4: SNP-Poor Regions "Plotted" on Chromosome Map
|
To include these on the chromosome map, I duplicate the locations for both paternal and maternal sides on my spreadsheet and in the column for MRCA (most recent common ancestor), I indicated "SNP-poor region", which then appears as such in the legend. These regions are shown as gray bars in Figures 1 and 2.
References
[1] Li H, Glusman G, Hu H, et al. (2014). Relationship estimation from whole-genome sequence data. PLOS Genetics 2014 Jan 30;10(1):e1004144 (http://dx.doi.org/10.1371/journal.pgen.1004144).
[2] Turner A (2010). Up hill and down dale in the genomic landscape: the odd distribution of matching segments. Journal of Genetic Genealogy 6(1).
[2] Turner A (2010). Up hill and down dale in the genomic landscape: the odd distribution of matching segments. Journal of Genetic Genealogy 6(1).
