Chromatin 'programming' by sequence--is there more to the nucleosome code than %GC?

Abstract

The role of genomic sequence in directing the packaging of eukaryotic genomes into chromatin has been the subject of considerable recent debate. A new paper from Tillo and Hughes shows that the intrinsic thermodynamic preference of a given sequence in the yeast genome for the histone octamer can largely be captured with a simple model, and in fact is mostly explained by %GC. Thus, the rules for predicting nucleosome occupancy from genomic sequence are much less complicated than has been claimed. See research article http://www.biomedcentral.com/1471-2105/10/442.

Figure 1

In vitro reconstitutions highlight yeast promoter nucleosome depletion. In vitro reconstitution data from Kaplan et al. [10] are shown in pink; data from our own in vivo nucleosome mapping [13] are in blue for comparison. Deep sequencing reads were mapped to the S. cerevisiae genome and extended to 140 bp (so each short read was extended to nucleosome length). Data were normalized for sequencing depth, and data for around 5,000 genes with well-defined transcriptional start sites (TSSs) were aligned and averaged over all genes for each dataset. Notably, the nucleosome depletion at yeast promoters is visible as a prominent valley in both datasets, whereas the stereotyped positioning of the +1 nucleosome relative to the transcription start site is clearly visible as a prominent peak only in the in vivo data. Red and green rectangles indicate regions previously proposed to be enriched for anti- and pronucleosomal sequences, respectively.