We have developed algorithms and programming tool set for genetic analysis of quantitative complex traits using extended recombinant inbred intercrosses and F2 population of their parental strains. Distinctive feature of the developed system is utilization of multi-QTL approach and taking account of parewise interloci interactions.
The problem solution includes several steps: identification a set of solutions able to account for extended recombinant intercrosses data up to the environmental noise; determination of loci genome positions; checking the similarity between the predicted and observed trait distributions using F2 population data; identification of solutions that explain joined experimental data; carrying out genome-wide permutation test for the joined datasets. We have developed and implemented multithreaded algorithms for the above steps of the approach.
Software deployment is supposed on the modern personal computers and laptops. The system is designed to suit multi-core processor architectures. We have achieved a significant increase in performance due to the parallel computing. Programming tool set was tested on real data with the cerebellum weight of laboratory mice being used as a quantitative trait.
