Linkage analysis in a large family from Pakistan with depression and a high incidence of consanguineous marriages.

Objectives: A genome wide scan for linkage was performed in a five generation family with a high incidence of depression and high average coefficient of inbreeding ascertained in a rural area of Pakistan. The effect of inbreeding on linkage analysis in an extended pedigree is discussed.

Methods: 372 microsatellite markers were used in a genome wide linkage study. Inbreeding coefficients were measured by two methods using both genealogical and genotype data.

Results: Of 111 family members with phenotypic information, 82 were diagnosed with recurrent major depression. Linkage analysis using the program Superlink online generated LOD scores of less than one at all loci. A model free analysis with SimWalk did not result in any significant linkage score. The mean inbreeding coefficient was 0.038 estimated from genealogical data and 0.02 estimated from the genotype data. These results did not differ significantly. The effects of inbreeding included a reduction in the polymorphism information content of markers and an overestimate of marker allele frequencies.

Conclusion: The analysis of very large families is computationally demanding. Problems encountered in this analysis, including loss of power due to reduced polymorphism information content and sensitivity of the LOD score method to estimates of allele frequencies, severely limited the chance of detecting linkage.