This method estimates the probability of sharing alleles identity by descent (IBD) accross the genome and can also be used for mapping disease loci using distantly related individuals. These individuals will often be simingly unrelated but if they share the same founder mutation then they will be distantly related. The method is based on a continouos time Markov model with hidden states. The hidden states are the IBD states between a pair of individuals with diploid chromosome. We assume that the individuals are not inbreed and thus the individuals can share 0, 1 or 2 alleles IBD. The SNPs are allowed to be in linkage disequilibrium (LD). To accommodate LD the methods need SNP for several individuals in order to estimate the allele frequencies and the pairwise LD. The method return the posterior probabilies of the IBD states across the genome and the overall IBD sharing. The estimates for all pairs of individuals can be combined info a score that will show linkage peaks across the genome and using a permutation procedure a significans threshold can be set. I recommend using the R package for fast visualization of a single pair of individuals (see figure).

Figure of the local relatedness on chromozome 13 for a sib pair estimated using affy 500k data
Figure of the local relatedness on chromozome 13 for a sib pair estimated using affy 500k data

Documentation updates (July 8 2009 )

There is minor typo in the manual on page 12. In sentence " If centi Morgan is used then phi should be set to 1." phi should be set to 0.01 instead.

Software (version 0.995)

So far this implementation is supplied as an R package for both windows and unix.

The method is implemented in an R package and as a commandline based C++ program embeded in the R package. The R code can be used to find and visualize the tracts of relatedness between a pair of individuals. The commandline version has under 20% of the running time when running all pairs compared to a single pair, it however has the the same speed for running a single pair analysis. For analysis linkage only the C++ version is implemented.

General R-package containing the full source can be downloaded here. This should be used if you want to compile the program for commandline usage.

There is also a precompiled R-package for windows that can be downloaded here. (Coming soon)

To compile the C++ version

See the manual.pdf for in-depth information about installation, method and examples.

If you have any problems or comments please contact me ande


The publication for the method is available from genetic Epidemiology:

Anders Albrechtsen, Thorfinn Sand Korneliussen, Ida Moltke, Thomas van Overseem Hansen, Finn Cilius Nielsen, Rasmus Nielsen. Relatedness mapping and tracts of relatedness for genome-wide data in the presence of linkage disequilibrium. Genetic Epidemiology

Change log

Version 0.995

Removed some NAMESPACE stuff that did not work on 2.13.0

Version 0.993 (3 april 2010)

added another linkage example with plot

Version 0.992 (18 jan 2010)

made compatable with gcc 4.4.1

Version 0.99 (12. april)

version 0.987 (24. feb 2009)

version 0.95

version 0.83

A better manual and more test files For download click here

version 0.802

apparantly the uint datatype doesn't exist in older compilers, So now a macro is added that defines the unsigned int if the gcc is older than 4.3