The structure of selective dinucleotide interactions and periodicities in D melanogaster mtDNA.

Carlos Y Valenzuela

doi:10.1186/0717-6287-47-18

Abstract

BackgroundWe found a strong selective 3-sites periodicity of deviations from randomness of the dinucleotide (DN) distribution, where both bases of DN were separated by 1, 2, K sites in prokaryotes and mtDNA. Three main aspects are studied. I) the specific 3 K-sites periodic structure of the 16 DN. II) to discard the possibility that the periodicity was produced by the highly nonrandom interactive association of contiguous bases, by studying the interaction of non-contiguous bases, the first one chosen each I sites and the second chosen J sites downstream. III) the difference between this selective periodicity of association (distance to randomness) of the four bases with the described fixed periodicities of base sequences.ResultsI) The 16 pairs presented a consistent periodicity in the strength of association of both bases of the pairs; the most deviated pairs are those where G and C are involved and the least deviated ones are those where A and T are involved. II) we found significant non-random interactions when the first nucleotide is chosen every I sites and the second J sites downstream until I = J = 76. III) we showed conclusive differences between these internucleotide association periodicities and sequence periodicities.ConclusionsThis relational selective periodicity is different from sequence periodicities and indicates that any base strongly interacts with the bases of the residual genome; this interaction and periodicity is highly structured and systematic for every pair of bases. This interaction should be destroyed in few generations by recurrent mutation; it is only compatible with the Synthetic Theory of Evolution and agrees with the Wright’s adaptive landscape conception and evolution by shifting balanced adaptive peaks.Electronic supplementary materialThe online version of this article (doi:10.1186/0717-6287-47-18) contains supplementary material, which is available to authorized users.

Highlights

We found a strong selective 3-sites periodicity of deviations from randomness of the dinucleotide (DN) distribution, where both bases of DN were separated by 1, 2, K sites in prokaryotes and mtDNA
For each series of the total dinucleotides with K separation we study the expected (Exp) and observed (Obs) number of the 16 possible pairs and obtain the stochastic continuous random variable known as the χ2 value [sum of (Obs-Exp)2/Exp] and the selection coefficient of each pair estimated by [(Obsi-Expi)/Expi; i from 1 to 16] expressed with positive and negative values
It is important to remark that our aim was never searching for sequence periodicities; on the contrary the method destroys them. We found among these selective interactions, as serendipity discovered by undergraduate students, a significant periodicity of the χ2 value for non-random dinucleotide bases’ association in prokaryotes and mtDNA [1,4,5,6,7,8,9]

Summary

Introduction

We found a strong selective 3-sites periodicity of deviations from randomness of the dinucleotide (DN) distribution, where both bases of DN were separated by 1, 2, K sites in prokaryotes and mtDNA. Our exigency of independence of DNA structures is satisfied because the first nucleotide (that one located upstream) runs () over all the sites of the chromosome excepting that one of the second nucleotide, and the second nucleotide (that one located downstream) runs over all the chromosome sites excepting that one of the first nucleotide, for each independent series of pairs (remark that nucleotides are re-sampled for each K-series). For each series of the total dinucleotides with K separation we study the expected (Exp) and observed (Obs) number of the 16 possible pairs and obtain the stochastic continuous random variable known as the χ2 value [sum of (Obs-Exp)2/Exp] and the selection coefficient of each pair estimated by [(Obsi-Expi)/Expi; i from 1 to 16] expressed with positive and negative values. The collections of pairs of each K series may begin at different sites and proceed up or downstream or both alternately and the result is the same, even though the samples of pairs do not include their total

Objectives

Methods

Results

Conclusion