Colloquium: “Statistical approaches to gene mapping”

Marshall University Math Colloquium
January 23, 2004

Yulia Dementieva
Marshall University

Statistical analysis methods for gene mapping originated in counting recombinant and non-recombinant offspring, but have progressed to sophisticated approaches for the mapping of complex trait genes. The current statistical methods used in genetic mapping are model-dependent LOD score analysis, model independent sibpair analysis and association study. In genome-wide mapping studies, depending on the stage of process, a particular method could be chosen as a complement to others and the best overall design may include multiple techniques. For example, sibpair analysis may be used in the initial screening but association study is preferred for detailed analysis of specific genomic regions. In this talk, the speaker will present various approaches to gene mapping and discuss their statistical challenges.