The persistence exponent of DNA.

Area of Science:

• Genomics
• Computational Biology
• Statistical Physics

Background:

• DNA sequences exhibit complex patterns beyond simple base pairing.
• Understanding base distribution is crucial for genomic analysis and evolutionary studies.

Purpose of the Study:

• To investigate the distribution of Guanine-Cytosine (G-C) content in DNA sequences.
• To quantify statistical persistence in DNA composition using a model of fractional Brownian walk.
• To compare G-C distribution patterns between a microbe (Thermoplasma volcanium) and a segment of the human genome.

Main Methods:

• Analysis of complete genome of Thermoplasma volcanium and a 10 million base segment of the human genome.
• Examination of distribution functions for G-C content in consecutive, non-overlapping DNA blocks of varying sizes (m).
• Application of Mandelbrot's fractional Brownian walk model to analyze statistical persistence and fractal dimensions.

Main Results:

• G-C content distributions are significantly broader than random distributions.
• A power law relationship was observed between block size and the width of G-C distributions.
• Statistical persistence of composition was identified, where blocks of similar G-C content tend to follow each other.
• Persistence exponents (gamma) were calculated: 0.29 for T. volcanium and 0.39 for the human genome segment.

Conclusions:

• DNA base composition displays significant statistical persistence, deviating from random models.
• The fractional Brownian walk model effectively describes this persistence and allows for scaling of G-C distributions.
• The strength of persistence, measured by the exponent gamma, shows similarities but also differences between microbial and human DNA.