In contrast, HBCGM divides the mouse genome into discrete regions, which are based upon the extent of linkage disequilibrium among all identified SNPs in a region. A new haplotype block is produced within a region when there is a change in the pattern of genetic variation among the inbred strains. A genetic map that accurately reflects the fine structure of genetic variation has multiple advantages over representative SNPs (Figure 1): (1) the dimensions of and strain groupings within each region are based upon complete knowledge of the pattern of genetic variation; (2) since all genetic variation is analyzed, the causative SNPs are included in the haplotype map; (3) the small sizes of the correlated genomic regions enable the formulation of a precise hypothesis about how a genetic variant could impact a trait value. For example, a haplotype block affecting H2-Eα gene expression was only 1 kb in size [7], which enabled an allelic effect on gene expression to be quickly characterized. If a phenotype is entirely determined by alleles at a single SNP, then GWAS and HBCGM are methodologically equivalent and will have the same detection power.
