Work

A Special Class of Oligonucleotides — Compound Events: Mathematical Properties of V4 and Structural Analysis of Human DNA

Year: 2026
Type: article
Abstract: A Special Class of Oligonucleotides — Compound Events: Mathematical Properties of V4 and Structural Analysis of Human DNA This paper investigates the structural organization of DNA sequences using compound events (CE) within the framework of Long Sequence Combinatorics (LSC). A compound event is defined as a maximal uninterrupted run of identical symbols. In DNA, compound events correspond to maximal homopolymer runs and represent a special class of homogeneous oligonucleotide sequences characterized by the additional condition of maximality. The study examines the unique decomposition of nucleotide sequences into non-overlapping compound events and establishes their mathematical relationship with homogeneous oligonucleotides (k-mers). The mathematical properties of an ideal random sequence consisting of four independent and equally probable symbols (V4) are considered. Theoretical expressions are presented for the total number of compound events, the number of compound events associated with each nucleotide type, and their distributions by run length. The theoretical predictions are compared with experimental compound-event counts obtained from 47 chromosome records representing two human reference genome assemblies: GRCh38/hg38 (23 chromosomes) and GRCh37/hg19 (24 chromosomes). The analysis reveals that the observed total number of maximal homogeneous nucleotide runs is systematically lower than the theoretical V4 prediction in all 47 chromosome records. The aggregate deviations are −6.621% for hg38 and −6.666% for hg19. An additional experimental analysis reveals a pronounced difference between two nucleotide groups in their deviations from the theoretical V4 model. For the A+T group, the aggregate compound-event counts are approximately 11% lower than the theoretical predictions. For the C+G group, the aggregate compound-event counts are much closer to the theoretical values, with deviations of approximately −0.5%. More specifically, the observed deviations are: GRCh38/hg38: A+T = −10.897%; C+G = −0.452%. GRCh37/hg19: A+T = −10.837%; C+G = −0.641%. The deviations for A and T are similar in magnitude, as are those for C and G. However, the differences between the two nucleotide groups are substantially greater. These findings establish an experimentally observed distinction between the normalized compound-event counts of A/T and C/G nucleotide groups relative to the ideal V4 model. The approximate values of −11% and −0.5% characterize the aggregate distributions across the analyzed chromosomes and are not universal constants for individual chromosomes. The interpretation considers the unequal nucleotide composition of real DNA, possible dependencies between adjacent nucleotides, and the fact that hg19 and hg38 are different assemblies of the human reference genome rather than independent biological samples. The results provide a quantitative basis for further investigation of the structural organization of DNA and the application of Long Sequence Combinatorics to bioinformatics. Supporting experimental datasets: https://doi.org/10.5281/zenodo.24381203 Additional calculations: KDP_Chargaff_CC_V4_pair_comparison.xlsx. Author: Oleg V. Filatov. Publication note: This publication is the English-language version of the original Russian research paper. The updated version includes an additional analysis of nucleotide-group-specific compound-event deviations from the theoretical V4 model.
Language: English
Open Access status: green
OpenAlex ID: W7237344516
FWCI:
Cites: 2
Cited by:
Related to: