[2] Genetic saturation occurs most rapidly on fast-evolving sequences, such as the hypervariable region of mitochondrial DNA, or in short tandem repeats such as on the Y-chromosome.
Divergence rates are estimated from a variety of sources including ancestral DNA, fossil records and biographical events.
[6] This use of molecular clocks to determine divergence is controversial because of its potential for inaccuracy and assumptions made in the model (such as consistent mutation rate for all branches) and is used mostly as an estimation tool.
[6] Genetic saturation can also be estimated by comparing the number of observed differences in nucleotide sequences between multiple pairs of species.
[2] In the field of molecular phylogenetics, the distances and relationships between species are investigated by looking at the DNA, RNA or amino acid sequences of an organism.
It can easily be confounded by genetic saturation because the homologous loci under investigation show no indication whether or not more than one substitution on each nucleotide separates the taxa being described.
The effects of saturation can mask the true amount of divergence time leading to inaccurate phylogenetic trees.
[1] The more substitution mutations, the more likely it is for previously dissimilar sequences to share nucleotides and as a result, show homology in phylogenetic tree calculations.
Long-branch attraction due to saturation has been proposed to be the cause of links in ancient phylogenies and puts into question even some of the earliest relationships between eukaryotes, archaea, and eubacteria.
[9][10] Researchers often lean towards using a one-step PCR-based to explore the specific effects of different variations in an amino acid of interest within a protein with GSSM.