Monday, October 21, 2019

Estimating Time Depth With Mutation Rates Is Intrinsically Complex

Early estimates of the time depth at which genetic clades emerged were based upon a fixed molecular clock model that turns out to not reflect reality. Mutation does happen at predictable rates, but a workable model is more complex. Not all kinds of genetic mutations take place at the same rate.
Michael E. Goldberg and Kelley Harris, Great ape mutation spectra vary across the phylogeny and the genome due to distinct mutational processes that evolve at different rates (October 15, 2019) doi: https://doi.org/10.1101/805598  
Recent studies of hominoid variation have shown that mutation rates and spectra can evolve rapidly, contradicting the fixed molecular clock model. The relative mutation rates of three-base-pair motifs differ significantly among great ape species, suggesting the action of unknown modifiers of DNA replication fidelity. To illuminate the footprints of these hypothetical mutators, we measured mutation spectra of several functional compartments (such as late-replicating regions) that are likely targeted by localized mutational processes. Using genetic diversity from 88 great apes, we find that compartment-specific mutational signatures appear largely conserved between species. These signatures layer with species-specific signatures to create rich mutational portraits: for example, late-replicating regions in gorillas contain an identifiable mixture of a replication timing signature and a gorilla-specific signature. Our results suggest that cis-acting mutational modifiers are highly conserved between species and transacting modifiers are driving rapid mutation spectrum evolution.
The authors also deserve credit for clearly packing the key ideas of their paper into the title. 

3 comments:

NeilB said...

Hi Andrew, did the authors give any hint that the mutation rate in humans should be faster or slower, than current theory? NeilB

andrew said...

It's open access on bioRXiv so you can read the whole thing. The emphasis is on the fact that a single mutation rate is just not a meaningful thing. The mutation rate in humans is higher on the first to PCAs than any of the other five species studied. So, the notion of a uniform mutation rate even for primate within the Great Ape clade does not hold up.

andrew said...

This language from the last two paragraphs of the conclusion suggests that using a data set trimmed to exclude portion of the genome more prone to variation could provide a better tool for serving as a molecular clock:

"In the majority of the genome where the replication timing landscape appears to be constrained, stable, and consistent in its mutational bias, we predict that the dosage of the replication timing signature is likely to behave as a more reliable molecular clock than the full spectrum of mutations that contains fast-evolving components and is known to vary in rate between lineages (Moorjani et al., 2016b, 2016a; Scally and Durbin, 2012). Using the spectrum loadings of the compartment signatures and species-specific signatures inferred here, we have enough information to estimate the number of mutations in an ape genome that were caused specifically by replication-associated signature. Given the apparent stability of this signature across ape species, this subset of mutations could be leveraged to estimate divergence times, population size changes, and other demographic parameters more accurately than can be done using mutations that accumulate preferentially in certain species.

If any mutational signatures accumulate at stable rates outside the clade of great apes, they could also prove useful for phylogenetic inference in larger sets of species as well. This has the potential to help resolve disputes about evolution and phylogeny in large clades where evolutionary inference is challenging. Although spatial and temporal variation of the mutation rate and spectrum both serve to distort the molecular clock model and complicate population genetic inference, their pattern of covariation suggests that spatially varying mutational signatures may provide the reliable molecular clock that has eluded evolutionary biologists for years."