Species-specificity And Pathogenicity Of Oligonucleotide Transition Probability Bias For Prokaryotic Genomes

As an important characteristicof genomes, oligonucleotide usage pattern and itsbiasreflects the whole individual genetic information atthe level of bases distribution, theexcessive or insufficient appearance of of which may lead tosome hereditary diseases.However, it remains unclear whether or not the bias of the genomic oligonucleotide usagepattern reflects the functions and species-specific properties of the genomes. Based on aMarkov chain model, a novel index, the characteristic vector of trinucleotide transitionprobability bias (tri-TPB), namely the pattern of maximum trinucleotide transition probabilitybias, was proposed to study the pathogenicity and species-specificity of prokaryotic genomes.The main contents include the following four aspects:(1) In order to identify the similarities of genomic composition bias and its relationshipwith G+C content, correlation analysis was used to compare characteristics tri-TPBcharacteristics vectors of727representative prokaryotic genome sequences. For the first time,our results showed that the closer the genetic relationship was, the more similar thecharacteristic of tri-TPB vectors was; especially, an identical maximum tri-TPB pattern wasinvolved nearly in all genomes within the same species, even in those with different genomeG+C contents.(2) Plasmids, phages and horizontally acquired genes are significant to the evolution ofprokaryotic genomes. Linear regression analysis of characteristic tri-TPB vectorsdemonstrated that the tri-TPB distribution of these exogenous gene elements was more similarto their host genomes than to other evolutionary distant strains, and the average correlationcoefficient decreased along with increasing evolutionary distance.(3) To examine the transition probability bias in virus with different hosts, characteristictri-TPB vectors of33dsDNA and dsRNA virus were studied. It was shown that the tri-TPBdistribution of viral genomes was not species-specific, and there was no relationship betweenthe difference of tri-TPB vectors and the host tropisms, which were completely different fromcharacteristics of the prokaryotic genomic tri-TPB distribution.(4) The highly correlation among tri-TPB vectors of closely related bacteria with similarpathogenicity suggested that the pathogenicity of closely related organisms might be related to their genomic maximum tri-TPB pattern. But this correlation is usually beyond or evenmisinterpreted by traditional phylogenetic analysis based on SSU rRNA sequences.In conclusion, an effective method based on maximum trinucleotide transitionprobability bias was proposed to sucessfully describe the species-specificity and pathogenicityof tri-TPB vectors of prokaryotic genomes, which might provide new ideas for future analysisof genome evolution as well as their pathogenicity evolution.