A method for estimating the number of invariant amino acid coding positions in a gene using cytochrome c as a model case

This paper shows, within the limitations of the assumption stated below, that approximately 27–29 of the unmutated codons which determine the amino acids of cytochrome c are invariant because of biological requirements. A mutation is defined here as the change of a single base in the sequence of a trinucleotide codon, which change alters the amino acid coded for. Codons, if any, in which mutations would be vigorously selected against are termed invariant codons. We assume that, subject to one adjustment, those mutations in the cytochrome c gene which survived in the descent of today's species are randomly distributed among the variable codons. The one adjustment arises from the possibility that a very few codon positions may exhibit frequencies of mutation sufficiently great to justify the exclusion of these codons from the overall distribution on the grounds that the frequency of mutation occurring in these few positions is clearly inconsistent with the assumption of randomness. There are 5 out of the total 110 codons in the cytochrome c structural gene which have clearly sustained an abnormally large number of mutations.This project received support from grants to W.M.F. from the National Institutes of Health (NB-04565) and the National Science Foundation (GB-4017).