In genome era,average nucleotide identity(ANI)is the "gold standard" of prokaryotic taxonomy to delineation of species.ANI is a mean of nucleotide identity between conserved regions shared by two genomes.Sincethe introduction of a DNA-DNA hybridization(DDH)value of 70%as the threshold for the microbial species boundary,95-96%of ANI values which correlate well with the thresholdis a robust measure of evolutionary distance.There is an issue,however,which difference between identities of query genome to reference genomeeould be a serious problem in the calculation of the mean values.Weinvestigate average nucleotide identities of 63,690 genome pairs with 14,745 strains in 10 big Genus.Our study found over 55%of total samples got more than 0.1%difference,and up to 1,101 genome pairs got difference not less than 1%.We develop a new method that similar with ANI algorithm to calculate identity between sets of truly orthologous sequences shared by two microbial genomes.In our study,we introduce modified ANI algorithm as name of OrthoANI.Average identity of nucleotide calculated via new algorithm hadone ANI value through tested version that had a 0.00042%average discrepancy between bidirectional calculations.