iSource is a program for source attribution (population assignment) using multi-locus genotypes with linkage disequilibrium. iSource can be used to assign sequences of unknown origin to putative source populations based on DNA sequence similarity to a panel of reference sequences of known source. It was specifically designed with multi-locus sequence typing (MLST) in mind, but can be applied more widely.

Source attribution probabilities are calculated using a statistical model of the way the DNA sequences evolve over time. In the model, there are parameters representing the processes of mutation, DNA exchange (recombination) and migration between populations. In the case of infectious disease, migration might correspond to transmission between host populations. These processes lead to differences in gene frequencies between populations, facilitating source attribution. The model is trained by estimating the evolutionary parameters exclusively from sequences of known source, before using it to calculate source attribution probabilities for sequences of unknown origin.

iSource differs fundamentally to similar population genetic analyses, notably Structure, in that it approximates a specific genetical model known as the migration matrix model, a generalization of Wright's symmetric island model. By modelling linkage disequilibrium between loci in the source populations, and by estimating the relative contribution of each putatitive source population to the sequences of unknown origin, iSource improves its accuracy of source attribution. This sensitivity is particularly important in weakly differentiated populations.

To download iSource click here. The zip file contains

Information on how to format your data for use with iSource, running the program, and example analyses are contained in the file README.txt. If you use iSource, or the example datasets, please cite the following paper.

Wilson, D. J., E. Gabriel, A. J. H. Leatherbarrow, J. Cheesbrough, S. Gee, E. Bolton, A. Fox, P. Fearnhead, C. A. Hart and P. J. Diggle (2008)
Tracing the source of campylobacteriosis.
PLoS Genetics 4: e1000203. (abstract pdf press-release)