Mathematical characterization of Chaos Game Representation. New algorithms for nucleotide sequence analysis

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Chaos Game Representation (CGR) visually identifies patterns in nucleotide sequences using fractal structures.
• Current CGR pattern recognition lacks a mathematical characterization, relying solely on visual identification.
• Understanding DNA sequence patterns is crucial for gene family classification and function prediction.

Purpose of the Study:

• To develop mathematical algorithms for characterizing Chaos Game Representation (CGR) patterns in DNA sequences.
• To provide a computational basis for predicting the presence or absence of nucleotide stretches within gene families.
• To establish a quantitative method for analyzing CGR patterns derived from nucleotide sequences.

Main Methods:

• Development of two algorithms: one to generate DNA sequences from CGR points, and another to simulate known CGR patterns.
• The second algorithm utilizes trial-and-error with guidelines to set di- and trinucleotide probabilities for CGR simulation.
• Algorithm validation involved simulating CGR patterns for vertebrate non-oncogenes, proto-oncogenes, and oncogenes.

Main Results:

• The algorithms provide a mathematical framework for understanding and generating CGR patterns.
• Successful simulation of CGR patterns for different gene families, including oncogenes and non-oncogenes.
• Demonstrated ability to predict nucleotide sequence patterns computationally, moving beyond visual identification.

Conclusions:

• The developed algorithms offer a mathematical foundation for Chaos Game Representation in DNA sequence analysis.
• These computational tools can aid in the identification and classification of gene families based on sequence patterns.
• The study bridges the gap between visual CGR pattern recognition and quantitative mathematical analysis.