The differences between proteins can be revealed and estimated using gel electrophoresis, a technique invented before the genetic material itself became accessible for direct study. Techniques applied to DNA directly are now more commonly used.
The paired strands of DNA dissociate when heated because the bonds between corresponding nucleotides are broken. They recombine on being cooled. Single strands of DNA from related species can be put together and bonds will form, but only at the sites which correspond. When such a ‘hybrid’ is carefully heated it dissociates at a temperature lower than that required to dissociate perfectly matched DNA, because fewer bonds will have been formed. The difference between the two temperatures (‘melting points’) can be used as a measure of the genetic similarity of the two species.
Sequencing of nucleotides
This process allows the identification of each nucleotide in the whole sequence of a DNA molecule. This exhaustive process has been made easier by the polymerase chain reaction (PCR), which amplifies a small quantity of material for rapid analysis, and by the automated sequencing machine. This technique opened the way to the present explosion of molecular information.
How is molecular information processed?
Molecular evidence provides a large number of characters, all precisely defined. A character is usually a given nucleotide at a given site on the DNA molecule. Can the methods used for morphological characters be applied?
The traditional method of assessing characters is clearly not applicable.
Phenetic analysis can be applied to molecular differences. If molecular change increases at a constant rate as evolution proceeds, the amount of change is a measure of evolutionary distance. However, genes do not always change at a constant rate (see below).
Cladistic analysis is generally used, and some of its drawbacks disappear, because the characters are precisely defined and equivalent, and sufficiently numerous for statistical analysis to be substituted for parsimony. Selection of an outgroup to root the cladogram is, however, even harder.