Figure 9: Polycodon Frequency vs. Codon Structure. Nucleotide Sequence Databases (NCBI) were searched by 64 different queries each containing 10x repeats of the 64 possible codons and the BlastN matches were automatically counted (default server settings). The collected data were divided into six groups corresponding to the composition of the 1st and 3rd codon positions (x axis) and the 2nd codon position (colors). Each column represents the mean ± S.D. of a total of n= 64 counting.