Construction Of Nucleosome Positioning Parameters Based On The Sequence Characters And Statistical Analysis Of Nucleosome Distributions In Transcription Start Site Regions In Human Genes

In the complex patterns of gene regulation in cells, transcription initiation is the first step to convert the DNA genetic information into proteins which is the basic material of cells. It was confirmed that the transcription start sites are normally lacking of nucleosomes, which is called nucleosome free region. Specific nucleosomes including+1nucleosome and-1nucleosome distribute on the flanking regions of the NFR. The special region of NFR provides binding sites accessible for transcription, and plays a key role in the regulation of gene expression. Experiments have demonstrated that the majority of transcription factor recruitment area is to avoid nucleosomes. This suggests that nucleosome positioning around the TSS is an important factor in transcription factor binding. Nucleosome positioning depends largely on the base composition of DNA sequence. Therefore, it is important for the regulation of transcription functions to study nucleosome positioning from the perspective of a sequence composing. Researchers have found that k-mer (k>6) frequency distribution of human intergenic sequences has tri-modal spectra, while its biological significance is still unclear. First of all, the three peak distribution of8-mer frequency were calculated by intergenic sequences of human chromosome1. The analysis revealed that if dividing the8-mers into three categories according to the8-mer containing larger or equal than2,1, and0CG dinucleotide, the three peaks can be completely separated. Analyses of sequence features in three8-mer subsets have suggested that containing1CG dinucleotide of8-mers is closely related to nucleosome binding motif. Based on the64relative frequencies of tri-nucleotides in ICG8-mer set, we selected15optimal and15rare tri-nucleotides to construct a characteristic parameter of nucleosome called Ktri to describe the information of nucleosome positioning. We calculated the Ktri distribution on the single segment surrounding transcription start sites (TSS) of human genes, and did the liner correlation between the nucleosome occupied rates which were obtained by experiment and Ktri values. The statistical results showed that there was a significant positive correlation between the two parameters. The similarity accuracy rate is89.2%(p<0.05) and81.6%(p<0.01) for the1177TSS segments. It suggests that the characteristic parameter of nucleosome is a sensitive quantity of nucleosome positioning, so the parameters can be used to predict the nucleosome positioning.In order to reveal the distributing character of nucleosome on the TSS region (-100bp～50bp), the Ktri quantity were used to classify and analyze58989segments around TSS of human genes. We observed that nucleosome can be combined to any position of TSS region, there are about28%genes in which±1nucleosome located flanking the TSS region denoted by the first class gene, and there are about30%sequences in which the nucleosome occupying TSS region denoted by the second class gene.The second-order information redundancy and GC content were used to analyze the base correlations and composition of the sequences including TSS. We noticed that the nucleosome core sequences have highly base correlation when the nucleosome without occupying TSS region. When the nucleosome occupy in the TSS region, the nucleosome core sequences have weak base correlation. No matter the nucleosome occupy the TSS region or not, the TSS regions always have weak base correlations. Thus, the weak base correlation is the common characteristics in TSS region. This result suggests that the nucleosomes flanking TSS have highly sequence adaptibility and position changeability. Traditional nucleosome free region of TSS is suitable only for the first gene class.In summary, based on the characteristics of frequency distribution of8-mer in human DNA sequence, we constructed a characteristic parameter of nucleosome which has a high correlation with the nucleosome occupancy rate. It is indicated that the characteristic parameter of nucleosome is a simple and effective theoretical quantity to predict nucleosome positions. We find that not all genes in the TSS region are absence of nucleosomes, nucleosomes may be located anywhere in the TSS region, and the base correlation in TSS region is always weak. The results suggest that the nucleosomes flanking TSS have more sequence adaptability. Therefore, we think that this kind of sequence adaptability is directly related to the gene expression level. This study has significance for the prediction of nucleosome positioning theoretically, and give a further understanding of the interactions between histones and DNA sequence in TSS region and mechanisms of expression regulation of gene transcription.