Distinguishing Functional DNA Words; A Method for Measuring Clustering Levels.

Area of Science:

• Genomics
• Statistical Mechanics
• Bioinformatics

Background:

• Genomic DNA exhibits spatial clustering of functional elements, analogous to word clustering in texts.
• Statistical mechanics offers methods to analyze these patterns, particularly non-extensive statistical mechanics.

Purpose of the Study:

• To adapt text analysis methods for comparing DNA sub-sequences.
• To introduce the q-parameter from non-extensive statistical mechanics as a measure of DNA sub-sequence clustering.
• To analyze dinucleotide clustering in human chromosomes and across species.

Main Methods:

• Analysis of the distribution of distances between consecutive dinucleotide occurrences in human chromosomes.
• Calculation of q-parameters for 16 possible dinucleotides.
• Comparative analysis of dinucleotide clustering across different organisms.

Main Results:

• The q-parameter effectively quantifies DNA sub-sequence clustering.
• The CG dinucleotide shows the highest clustering level, linked to its biological significance (e.g., methylation).
• Chromosome 18 exhibits high promoter q-parameter values, suggesting sensitivity to environmental factors.
• CG dinucleotide clustering increases with evolutionary complexity.

Conclusions:

• The q-parameter is a valuable tool for analyzing genomic sequence organization.
• High CG clustering indicates functional importance and evolutionary trends.
• Genomic clustering patterns can predict biological sensitivity and evolutionary relationships.