Histone modifications: from genome-wide maps to functional insights

Abstract

A large number of histone modifications has been implicated in the regulation of gene expression. Together, these modifications have the potential to form a complex combinatorial regulatory code. Genome-wide mapping approaches provide new opportunities to decipher this code, but they may suffer from systematic biases. Integration of datasets and improved technologies will provide the way forward.

Figure 1

Two models for the role of the different histone modifications. (a) Histone modifications are linked, such that all active genes are marked by an identical combination of 'active' histone modifications, and all inactive genes by another common set of 'repressive' histone modifications. (b) Histone modifications are not linked but form a complex combinatorial code, such that various combinations of modifications mark (groups of) active genes, and various other combinations mark inactive genes. Note that some organisms (such as budding yeast) may lack 'repressive' histone modifications.

Figure 2

Effect of nucleosome distribution on chromatin immunoprecipitation. Uneven genomic distribution of nucleosomes (a) may cause a bias in the results of chromatin immunoprecipitation experiments (b). The bias may be solved by including ChIP controls for nucleosome density (see text) and subsequent normalization of the histone-modification ChIP data.

Figure 3

Comparison of the correlations of ChIP of each lysine acetylation with nucleosome density and with gene expression levels. Data are from Table 1.