How do we logically infer large time scales using Molecular Clocks?

Question

If molecular clocks are uses in genetics to determine the mutation rate of genes to estimate times speciation occurred between two or more life forms, are there genes that have mutation rates that we can observe that allows us to deduce mutation rates over hundreds of thousands to millions of years?

For example, Homo Sapien speciation occurred 300,000 years ago from Homo Erectus that existed 2 million years ago. How did we get those time scales? Are there mutation rates in simple bacteria or viruses that are constant that we can use to logically build up to those longer time scales?

Answer 1

There are a few different questions going on here.

1. Mutational clocks are not the only, or even the primary, means of dating specimens of these species. More common is to use carbon dating, for example as done here, to establish when certain species existed.
2. Mutational molecular clocks are calibrated based on observed mutation rates. It is fairly routine to measure mutation rates per generation using whole-genome DNA sequencing. We have a good vision of what substitution rates are on average throughout the genome.

For an intuition of why this works, consider that there are roughly $3*10^9$ bp in a haploid human genome. The human mutation rate is about $1*10^{-8}$ per generation. So every generation, we expect around ~30 new mutations on average per haploid genome. It's simply a matter of observing the whole genome to find them.

Mitochondrial genome mutation rates are quite a bit faster, and were some of the first genomic regions used to do this kind of dating, before whole genome sequencing became so routine.

Using that kind of information, we can make some statistical guesses about the relationships between Homo species (I'll let you look at some of those references to see how that is done).