Scaling behaviors of CG clusters for chromosomes

In this paper we adopt a new method to study the scaling behaviors of CG clusters in different organism chromosomes. The statistical distributions of CG and AT clusters for different chromosomes have the same scaling behaviors, i.e. P(S) ∝ e−αS. The values of α are very close to each other for the same organism chromosomes, and depend on different organism chromosomes. We also find that the parameter ξ(m)=σ(m)m of CG cluster complies with the good power law ξ(m) ∝ m−γ. Here σ(m)=〈n¯〉2-〈n¯2〉, and m is the number of bases in consecutive, non-overlapping blocks. The values of γ have the same behavior as the values of α in statistical distributions of P(S) ∝ e−αS. Meanwhile, we also consider the relationship between the values of γ and the percentage of cluster CG content for different organism chromosomes, and there are some relations between them. These investigations provide some insights into the nucleotide clusters of chromosomes, and help us understand DNA sequences of chromosomes.