The goal of linkage analysis is to locate on the genome the gene (or genes) responsible for an inherited trait, typically a disease under study. The power of this method lies in the fact that it is not necessary to identify or, indeed, even know, the function of the gene of interest in order to estimate its location on the genome. This approach is also commonly known as reverse genetics.

While standard methods of linkage analysis have proved enormously powerful for relatively rare diseases with simple patterns of inheritance (e.g., Huntington's chorea, cystic fibrosis, neurofibromatosis, etc.), they do not fare so well in localizing genes predisposing to more common disease with complex patterns of inheritance, such as asthma or diabetes.

MLIP is a new version of the well-known LIPED linkage analysis code. This is a mixed-language effort (Fortran and ANSI C) that builds on our own work in distributed computing to calculate, in parallel, LOD scores by pedigree and marker over a user-specified range of putative disease gene frequencies, penetrances, and male and female recombination fractions. The output of MLIP can then be used to calculate either the posterior probability of linkage integral or the maximal LOD score over the parameter space.

This is joint work with statistical geneticists in the Center for Statistical Genetics Research.