Estimating Human Recombination through Computational Methods

Human genetic recombination can be mapped by segregation analysis; genotyping carried out on sperm and somatic cells, or by population based inferences using genetic markers e.g.- Single Nucleotide Polymorphisms (SNPs). The HapMap has provided a watershed of SNP data that can be used to estimate recombination among four populations. A series of algorithms developed over the past five years permits genome wide recombination estimates when sufficient computing power is applied. 
We used several methods to estimate local recombination rates on chromosomes 11-22, as well as the Major Histocompatibility Complex (MHC) at 6p21-22, the pseudoautosomal region of Xp, and a 1Mb region at 1q42.3. We externally validated our results through comparison to well–characterized recombination measurements published by others, and internally validated our work through comparison of the results of two different computational algorithms. Our recombination estimates recapitulate the well-known human TAP2 recombination hot spot and the strong marker association hot spot on chromosome 1. Our observations in the MHC are consistent with published data that suggests hot spots and cold spots are frequent genomic neighbors, and that Tokyo Japanese and Han Chinese recombination patterns bear great similarity to each other. Questions still remain about the reliability of population historic recombination estimates versus those obtained by mapping or sperm typing, and about whether there are intrinsic DNA sequence based structures that catalyze intense recombination and therefore are characteristic of hotspots.