(A) The location of the Z loci along Chromosome III is depicted by gray bars.
(B) A zoom in a 120 KB region between position 196611–316611 shows the raw H2A.Z/H2B log₂ ratios for each probe in that region (green bars) and the smoothed data resulted from the Gaussian plot analysis of the raw data for H2A.Z/H2B log₂ ratios (red line). The position of the Z loci is shown by gray lines. The genes present in that regions are shown in blue.
(C) Same as in (B) but for a zoom in region 237700–277700 (40 KB).
(D) The size of the Z loci within the promoter of the SRB8 gene as determined by Q-PCR analysis of ChIP experiments. The binding ratios for H2A.Z (red) and TFIIB (blue) are shown relative to the center of the probe that generates the maximum enrichment. The data were smoothed by a sliding median over three probes (see Figure S1). The size of the observed H2A.Z domain is about 250 bp larger than that of TFIIB. Since TFIIB covers about 10 bp of DNA, we can infer that H2A.Z covers about 260 bp of DNA at that locus (shaded area). More details can be found in Protocol S1 and Figure S1.

(A) A sliding median of the H2A.Z/H2B ratio for all probes on the microarray is plotted against the distance from the 5′ boundary of their closest transcribed element (including ORF, tRNA, transposon, etc.) (green). Randomized data (where the H2A.Z/H2B ratios were scrambled prior to calculating the sliding median) are plotted in black.
(B) Most Z loci are found within promoters. The fraction of the 4,862 Z loci present within promoters (gold) and nonpromoter region (aqua) is shown. Promoter regions are defined as the −500/+100 interval relative to the 5′ boundary of transcribed elements as annotated in the SGD database.
(C) The fraction of the 7,571 promoter regions (defined as in [B]) containing (green) or not containing (purple) a Z loci is shown.
(D) H2A.Z is preferentially localized downstream of the NFR in yeast promoters. A sliding median of the ratio for H2A.Z/H2B (green) and H2A/H2B (gold) was plotted against the distance from the NFR (defined as a linker of more than 100 bp located less than 500 bp upstream of a 5′ gene boundary) for regions where the NFR can be unambiguously assigned to a single gene (i.e., NFR that map to two diverging genes were filtered out). A similar sliding window was applied to the nucleosome positioning data from Yuan et al. [37] (blue). A cartoon representation of the nucleosomal organization of a typical gene is shown at the bottom.

The binding trend was calculated by computing the moving median of the H2A.Z/H2B flat ratios over a sliding window of 50 genes across all genes ordered by transcription rate as described previously [36,40,41,60–62]. Each probe was assigned to its closest ATG, and the probe with the highest H2A.Z/H2B value for each gene was used. The transcription rate for yeast genes was determined previously [63].

(A) Nucleosome positioning map of genes associated with a Z locus. The data from Yuan et al. [37] were used to compute the nucleosome occupancy curve for all genes containing a Z locus aligned on their ATG. Peaks represent nucleosomes, and valleys represent linker regions. An NFR is detected approximately 200 bp upstream of ATGs as described by Yuan et al. [37]. The vertical thickness of the curves contains 1-SD error bars for the mean log₂ ratio.
(B) Same as (A) but for genes containing no Z locus.
(C) Indirect end-labeling of MNase-digested chromatin from WT and htz1Δ cells grown in the presence of glucose.
(D) High-resolution LM-PCR analysis of MNase digested nucB–C mononucleosomes. Upper part: structure of the GAL1 promoter and PCR probes used; left: nucB analysis probing the right (R), and left (L) boundaries in WT and htz1Δ (Δ) cells; right: same as left part of the figure, but the analysis is with nucC.

(A) The raw (green) and smoothed (red) H2A.Z/H2B log₂ ratios, as determined by our microarray experiment, are shown across a 5-KB region around the GIT1 genes. The genes present in that region are shown in blue.
(B) Same as (A) but for a 5-KB region around the SRB8 gene.
(C) H2A.Z covers wider regions near telomeres. The size of the regions covered by H2A.Z (as determined by distance between the coordinates where the smoothed H2A.Z/H2B log₂ ratio reaches “0.1”) is plotted against the distance to telomeres. A moving average (window = 10 KB) was applied to the data.