Abstract: Genome-Wide Association Study（GWAS） is an important method to search for susceptible genes of complex diseases, and practical and effective algorithms are in need in GWAS. Testing GWAS algorithm on simulation data based on different disease model is important in comparing their performances. Generating the simulation data requires calculating the parameters of diseases models according to the control parameters, but there are no corresponding public algorithms. This paper proposes a branch and bound algorithm to compute the parameters of diseases models. Extensive experimental results show that the algorithm works out quickly with accurate parameters of diseases models. The algorithm can be used in constructing test platforms for GWAS algorithm.