Distance is powerful in the sense that it can be used with anything that can be measured. For example, the distance could be based on the strength of an immunological reaction. Using this method any form or measure of distance can be used and different types of measures can be combined into one. Hence, distances can be used with restriction site data, with allozyme data, with data on quantitative characters, with DNA fingerprints or even with real finger fingerprints. Methods to correct this type of data are not well developed because these are not as well defined characteristics.
Even with amino acids, the corrections can not be done easily and/or without some large bias. A Jukes-Cantor correction is possible it is simply
or more commonly,
But this assumes (as does the nucleotide Jukes-Cantor correction) that for all characters the rate of substitution from one amino acid and to some other amino acid are equal and independent of the residue. This is not true of DNA and is even less true for proteins. Amino acids like cysteine and proline are very important for the structure and function of proteins. Amino acids such as tryptophan have bulky side groups and can not be inserted easily into any site in a peptide. Because of this most amino acid distances use empirical weighting schemes. The most popular of these empirical measures is the PAM family of matrices.