Top twenty 10 kb regions in the human genome by estimated TMRCA.

a<p>Genomic coordinates in hg19 assembly. The genome was simply partitioned into nonoverlapping 10 kb intervals in hg19 coordinates.</p>b<p>Posterior expected TMRCA in generations, averaged across unfiltered genomic positions in region.</p>c<p>Number of polymorphisms in Complete Genomics dataset in region per kilobase of unfiltered sequence.</p>d<p>Normalized polymorphism rate: number of polymorphisms per unfiltered kilobase divided first by the local mutation rate (as estimated from divergence to nonhuman primate outgroup genomes) then by the average of the same polymorphism/divergence ratio in designated neutral regions. The resulting value can be interpreted as a fold increase in the mutation-normalized polymorphism rate compared with the expectation under neutrality. The same measure was computed from the much larger 1000 Genomes Project Phase 1 data set, and was significantly elevated in these 20 high-TMRCA regions (Supplementary <a href="http://www.plosgenetics.org/article/info:doi/10.1371/journal.pgen.1004342#pgen.1004342.s013" target="_blank">Figure S13</a>).</p>e<p>Possible copy number variant (CNV), based on Complete Genomics “hypervariable” or “invariant” labels (see <a href="http://www.plosgenetics.org/article/info:doi/10.1371/journal.pgen.1004342#s4" target="_blank">Methods</a>). Polymorphism rates in these regions may be over-estimated.