Simple models of genomic variation in human SNP density

Area of Science:

• Genomics
• Population Genetics
• Bioinformatics

Background:

• Observed variation in single-nucleotide polymorphism (SNP) density across the human genome was analyzed.
• Descriptive hierarchical Poisson models and population-genetic coalescent mixture models were employed.

Purpose of the Study:

• To describe and model the heterogeneity of SNP density in the human genome.
• To estimate genomic heterogeneity in the scaled mutation rate (theta).

Main Methods:

• Utilized empirical estimates of human genome recombination rates.
• Applied maximum likelihood estimation to SNP density distributions.
• Compared models with and without recombination heterogeneities.

Main Results:

• Genomic heterogeneity in the scaled mutation rate (theta) was estimated using empirical recombination rates and SNP density.
• Models incorporating recombinational heterogeneities provided significantly better fits to observed SNP density than those that did not.
• The scaled mutation rate theta was found to be heterogeneous across the human genome.

Conclusions:

• Acknowledging mutational and recombinational heterogeneities is crucial for developing empirically sound null distributions in genome scans.
• This approach enhances the detection of evolutionary outliers by accounting for unobserved historical phenomena.
• Improved models allow for more robust inference in population genetic studies.