Information and order of genomic sequences within chromosomes as identified by complexity theory. An integrated methodology

TL;DR

This study introduces an integrated methodology combining complexity metrics, machine learning, and non-extensive statistical theory to analyze the size distribution and information order of genomic sequences within human chromosomes, revealing non-random, patterned complexity behaviors.

Contribution

The paper presents a novel integrated approach using complexity theory, Tsallis statistics, and machine learning to uncover the non-random, patterned complexity of genomic sequence distributions within chromosomes.

Findings

Intron regions exhibit higher complexity and longer-range correlations than exons.

Genomic size distributions follow specific, non-random patterns with characteristic complexity features.

DNA sequences show multifractal characteristics and long-range correlations, indicating complex dynamics.

Abstract

Complexity metrics and machine learning (ML) models have been utilized to analyze the lengths of segmental genomic entities like: exons, introns, intergenic and repeat/unique DNA sequences, in each of the 22 human chromosomes. The purpose of the study was to assess information and order that may be concealed within the size distribution of these sequences. For this purpose, we developed an innovative integrated methodology. Our analysis is based upon the reconstructed phase space theorem, the non-extensive statistical theory of Tsallis, ML techniques and a new technical index, integrating the generated information, which we introduce and named it Complexity Factor (COFA). The low-dimensional deterministic nonlinear chaotic and non-extensive statistical character of the DNA sequences was verified with strong multifractal characteristics and long-range correlations with significant variations per genomic entity and per chromosome. The results of the analysis reveal changes in complexity behavior per genomic entity and chromosome regarding the size distribution of individual genomic segment. The lengths of intron regions show greater complexity behavior in all metrics than the exonic ones, with longer range correlations, and stronger memory effects, for all chromosomes. We conclude from our analysis, that the size distribution of the genomic regions within chromosomes, are not random, but follow a specific pattern with characteristic features, that have been seen here through its complexity character, and it is part of the dynamics of the whole genome according to complexity theory. This picture of dynamics of the redundancy of information in DNA recognized from ML tools for clustering, classification and prediction.