IBDelphi Trio
IBDelphi Trio was developed to more accurately calculate the chance that both members of a consanguineous couple are carriers for a recessive autosomal disease which is known to occur in the family, but has not been mapped to a specific gene or region. While it is possible to determine a probability based on the pedigree structure, due to the manner in which chromosomes are inherited this figure could significantly over or under estimate the real value. IBDelphi Trio, works by initially comparing the genotype data for the consanguineous couple and then compares this to the autozygous regions in an affected relative. IBDelphi Trio then identifies what proportion of the affected relative's autozygous regions is found to have IBD in the couple. Consequently, IBDelphi Trio assumes that the disease gene is located in an extended autozygous region in the affected relative.
Data analysis
Analysing the data files
Figure 1
Genotype data files for each member of the couple and the affected relative is added by pressing the appropriate button in the upper panel of the main form. When these files have been entered, the button in the middle panel of the main form becomes active. Pressing the prompts the user to choose which positional data to use. Only the positional data in the males genotype data file is available. Once the distance units have been selected the data is imported and analysed. First IBDelphi Trio creates the SNP database, loads the couple's genotype data, followed by the affected relatives data and then scans the genotype data for regions IBD and autozygosity. When complete the three buttons in the lower panel become active (figure 2).
Figure 2
In situations where it is not possible to obtain genotype data for both members of the couple, it is possible for IBDelphi Trio to compare the genotype data from a single member of the couple and the autozygous regions of the affected relative. To identify which SNP genotype file is missing, tick the appropriate check box, below the individuals button. Omitting a member of the couple may result in a significant increase in the extent of IBD regions identified between the remaining member of the couple and the autozygous regions of the affected relative.
Viewing the results of the analysis
Viewing a summary of the analysis
To view a visual summary of the couples IBD regions with reference to the affected relative's autozygous regions, click the button in the lower panel. This opens a new window which displays the results for all the chromosomes (Figure 3). The black, gray and yellow lines represent SNPs that:
- Black: exclude the flanking sequences from having IBD in the couple.
- Grey: exclude the flanking sequences from having a common haplotye between the couple and affected relative
- Yellow: Do not exclude IBD.
Consequently, regions of possible IBD appear as blocks of yellow with the occasional black or grey line which suggests a miscalled SNP genotype. Non-IBD regions appear as a combination of black, gray and yellow lines, while white areas show regions with no SNP coverage. Each chromosome may be flanked by blue and red lines, these represents autozygous regions in the male (blue lines) and female (red lines) genomes. Similarly, the pale green blocks represent autozygous regions in the affected relatives genome. This image can be saved by pressing the in the lower left corner of the window.
Figure 3
Viewing a detailed display of IBD regions on each chromosome
The IBD and autozygosity data for each chromosome can be viewed by pressing the button; this opens a new window that displays the results for a single chromosome (Figure 4). The current chromosome is identified in the windows title bar and is selected using the drop down list in the lower left hand corner of the window. As in the previous window possible IBD regions are identified by an extended region of yellow lines with few, if any, black or grey vertical lines. Each vertical yellow line shows the position of a non-excluding SNP, where as vertical black lines identify SNPs that exclude possible IBD between the couple's genomes and grey lines exclude a common haplotype that is homozygous in the affected relative. If either of the male or female genomes contains extended regions of homozygosity they are shown as blue (male genome) or red (female genome) rectangles flanking the chromosome. While autozygous regions in the affected relative are shown as pale green rectangles.
Since it is assumed that the affected relative originated has parents, the disease is likely to be a caused by an autosomal recessive mutation located in one of its autozygous regions. Similarly, it is unlikely to be located in an autozygous region of one of the unaffected members of the couple. Consequently, only regions that have a common haplotype in the couples genomes, that is homozygous in the affected relative, but not in either of the couples genomes is regarded as important.
Figure 4
The default view of this window highlights regions that do not have a common haplotype in the couple that is homozygous in the affected relative. Unticking the check box highlights these regions and creates a view very similarly to IBDelphi (figure 5).
Figure 5
In consanguineous couples from outbreed populations it is unlikely IBD regions will occur on both copies of a chromosome pair. However in couples from an inbreed population it is possible for a region to have IBD from two distinct common ancestors. Ticking the check box displays a second chromosome in which extended regions of yellow suggest both chromosome copies show IBD (Figure 6).
Figure 6
Due to the limited screen resolution, compared to the large number of SNPs per chromosome, multiple SNPs are likely to occupy the same pixel on the screen. It is consequently difficult to discern whether a region has been excluded by just a few or by many SNPs. To give an indication of the number of SNPs that exclude a region, the window also shows a graph of the number of non-excluding SNPs in a sliding SNP window of 900 SNPs. Since most SNPs are uninformative, the graph only shows regions that have 25 or fewer excluding SNPs. If the check box is ticked the graph also shows a red curve which indicates the number of SNPs that exclude a regions from been a region of dual IBD (Figure 6). Since the graph shows the score across the sliding window of 900 SNP, the IBD region may project either side of the region indicated by the graph. Ticking the check box highlights the true extent of the predicted IBD and predicted dual IBD regions as a thick black and red line.
It is possible to save this window as an image file by pressing the button. Clicking the button creates a web page that contains images of possible IBD regions of interest on all the autosomal chromosomes, an example web page can be seen here. The format of each image depends on the options chosen using the check boxes at the bottom of the window.
Viewing a text summary of the possible IBD regions in a couple
Figure 7
Pressing the button on the main window opens a new window which summarizes the analysis (Figure 7). The summary includes the name of the SNP genotype data files used and the size of the autosomal genome covered by the SNP data. This value does not include the sex chromosomes or the regions with no SNP coverage in the data file such as the P arm of chromosomes 13, 14, 15, 21 and 22. The total length of IBD regions found in the couple is stated as both a physical size and as a proportion of the autosomal genome.
Since the disease mutation in the affected relative is probably in an autozygous region, the risk of both of the members of the couple been a carrier of the disease allele is given as the proportion of the total length of IBD regions in the couple that are homozygous for the same haplotype in the affected relative, but not homozygous in either of the couple, compared to the total length of homozygous regions in the affected relative.
Finally each region found is annotated with the chromosome it is on, its size, start and end points and the total size of any gaps found in the region. The gap size is important in IBD fragments that span regions of very low SNP coverage, such as the regions flanking the centromere on chromosomes 1 and 9. These gaps are automatically removed from the total size of IBD regions found. This text can be either copied and pasted to a new document or saved as a text file by clicking the in the lower left corner of the form.