From Sequence to Morphology - Long-Range Correlations in Complete Sequenced Genomes

The largely unresolved sequential organization, i.e. the relations within DNA sequences, and its connection to the three-dimensional organization of genomes was investigated by correlation analyses of completely sequenced chromosomes from Viroids, Archaea, Bacteria, Arabidopsis thaliana, Saccharomyces cerevisae, Schizosaccharomyces pombe, Encephalitozoon cunniculi, Drosophila melangoster, Homo sapiens, chloroplasts and mitochondria. All sequences revealed long-range power-law correlations almost on the entire observable scale. The local correlation coefficient shows close to random correlations on the scale of a few base pairs, a first maximum from 40-3400 bp, and often a region of one or more second maxima from 10^5-3x10^5 bp. This multi-scaling behaviour is species specific and can be explained by a block organization of genomes. Within this multi-scaling behaviour an additional fine-structure is present and attributable to the codon usage in all except the human sequences. Here it is connected to nucleosomal binding. Computer generated random sequences assuming a block organization, the codon usage and nucleosomal binding agree with these results. Mutation by simulated sequence reshuffling destroyed all correlations, thus their stability seems evolutionary tightly controlled and connected to the spatial genome organization. On large scales the sequence correlations agree very well with the three-dimensional folding of the 30 nm chromatin fibre into the Multi-Loop-Subcompartment (MLS) model, in which ~100 kbp loops form rosettes, connected by a linker, within chromosomes.