Conservative nature of mutant substitutions

It has been known, since the early work of Zuckerkandl and Pauling (1965), that in protein evolution, amino acids which are similar in structure and chemical properties are substituted more frequently than dissimilar ones. They considered that such ‘conservative’ substitutions are likely to cause only a small change in the function of the molecule, and are therefore more likely to be ‘adopted by natural selection’. They noted, however, that ‘apparently chemists and protein molecules do not share the same opinion’ regarding what are the most prominent properties of a residue.

So, they assumed that the best criterion for conservatism of a substitution is a high frequency of occurrence during protein evolution. One shortcoming of such a criterion is that arguments easily become circular.

It is desirable, therefore, to use a more objective measure of amino acid similarity, and then investigate whether there is a positive correlation between such similarity and the frequency of evolutionary substitution. Another point about which we have to be careful in such an analysis is that the genetic code has the property, already noted by Zuckerkandl and Pauling (1965), that a single base substitution often leads to a similar amino acid or to a synonymous codon.