摘要：In this study,a new strategy for estimating codon homozygosity when there is an albsence of amino acids was introduced．Totally 124 eubacterial genome codon usage data without absence of amino acids were used in our test．The actual and predicted Nc(aa)values of 18 multirpe-codon amino acids in those genomes were investigated．The results reveal that：(1) Nc(aa) of amino acids ofdegeneracy of three or four was more precisely predicted；and(2)codon usage bias of most amino acids of degeneracy of two or six Was overestimated．The analyzed results based on the current algorithm imply that Our method is a promising way for estimating codon homozygosity of the absent amino acid，and we propose it can be further used to evaluate the relative importance of nucleotide composition in determining codon usage．
