ABSTRACT
Most methods to identify natural selection using molecular data rely upon the assumption that molecules contribute a relative amount to the biological fi tness of individual organisms. This relative contribution is often difficult to quantify, although molecular evolution studies work on the assumption that essential proteins are generally subjected to stronger selective constraints than nonessential proteins. While this view can be challenged by the fact that shifts on selective constraints are possible from one group of organisms to another for a particular protein, essential proteins in one organism are generally so in its close phylogenetic relatives. How could we possibly identify the signatures of natural selection using molecular data? In a previous chapter, we have learned that protein-coding genes comprise sets of codons (nucleotide triplets), in which mutations have different rates of fi xation
depending on whether they fall in the fi rst, second or third positions of the codon. Mutations that lead to amino acid replacements, generally those falling in the fi rst and second codon positions, are generally more harmful than the ones that do not change the amino acid. However, the relative importance of amino acid substitutions can differ according to the implication of the affected amino acid in the protein function or structure. Amino acid replacements that do not alter any of the properties of proteins, hence having no effect on the performance of an organism, are known as neutral changes. Those replacements that involve a change in the protein’s function are subjected to the fi lter of natural selection.
