Figure 4.

Nucleosome fuzziness relative to TSS. (a) Fuzziness is reported as the standard deviation of the six input nucleosome locations for each individual reference nucleosome. Nucleosome distances from the TSS were binned in 10-bp intervals, and the distribution smoothed using a three-bin moving average. Nucleosomes were required to have at least 50% occupancy and be called by at least four of the datasets. (b) Illustration of a nucleosomal array and NFRs (≥147 bp) with particular emphasis on border nucleosomes at the 5' end of genes (+1 position) in comparison with those elsewhere in the genome (that is, not at positions -1 through +4, nor at the end of genes nor in intergenic regions; nucleosomes were required to have ≥50% occupancy and be called by at least five datasets). Shown are bar graphs of quantitative measures of nucleosome fuzziness, H2A.Z/H3-H4 ratios, and poly dA:dT (A_≥5 or T_≥5) density in all nucleosomes or NFRs (≥147 bp) having the illustrated property (border versus non-border nucleosomes, and 5' NFR versus genic NFRs).