Fig. 27. Comparison of frequencies of two neighboring amino acid residues observed on protein sequences from bacterial genomes with those estimated by calculation with amino acid compositions of proteins encoded by genomes (H. influenzae (lowere plots) and H. pylori (upper plots)). Points of 400 combinations of two neighboring amino acids should distribute around the linear line with a slope of 1, if proteins were produced through random peptide formation of amino acids restricted in protein-0th-order structures, such as and GC-NSF(a)-0th-order structures. Upper plots H. pylori proteins were shifted by addition of 0.008 to the raw data to make two sets of data to compare easily.