FounderTracker is a computational method for detecting significantly recurrent identity by descent (IBD) in a set of haplotypes. This tool was designed for the discovery of founder mutations in dense SNP array data from a population sample of tumors (Letouzé et al., submitted).

In brief, the program uses the GERMLINE (Gusev et al., 2008) algorithm to detect pairwise IBD in the set of haplotypes. Each pairwise IBD segment is assigned a score based on its length and allelic content. An IBD score is then calculated for each SNP marker, by integrating the scores of pairwise IBD segments. Finally, a p-value is derived for each SNP by comparing its IBD score in tumors with a null distribution obtained from simulated data sets generated with the HAPGEN software (Marchini et al., 2007).

Using FounderTracker
FounderTracker can be used online at the FounderTracker web server. A standalone version can also be downloaded here.

An example of input data corresponding to 13 childhood adrenocortical tumors can be downloaded  here.

FounderTracker requires 4 input files:
Note that these formats correspond to the input formats for IMPUTE software (Marchini et al., 2007). Reference haplotypes from HapMap data can be downloaded, together with the legend and recombination files, from the IMPUTE web page.

Five parameters must be specified:
Note that the parameters min_m, err_hom, and bits are parameters for the GERMLINE algorithm, used in the first step of the method.

FounderTracker generates 5 output files:
The Integrative Genomics Viewer can be used for interactive visualization of FounderTracker scores, tumor haplotypes and genes within conserved haplotypes (see figure below).

Visualization with the Integrative Genomics Viewer of conserved haplotypes on the 17p chromosome arm in 13 childhood adrenocortical tumors .

If you use FounderTracker in a publication, please cite the article describing FounderTracker as well as the papers describing the GERMLINE and HAPGEN algorithms used in our method.

FounderTracker: Letouzé et al., submitted.

GERMLINE: Gusev A, Lowe JK, Stoffel M, Daly MJ, Altshuler D, et al. (2009) Whole population, genome-wide mapping of hidden relatedness. Genome Res 19: 318-326.

HAPGEN: Marchini J, Howie B, Myers S, McVean G, Donnelly P (2007) A new multipoint method for genome-wide association studies by imputation of genotypes. Nat. Genet 39: 906-913.

If you have any questions or comments please contact the developers at: