This isn't my area particularly, but of course generation rate will play a major role. Many terrestrial and shallow marine arthropods can have several generations per year (especially in warmer climates), and one generation per year is the normal. This is substantially more than most (at least larger) vertebrates.

So, I would generally expect that invert mutation rates are higher than vertebrates, but it will be different between lineages depending on various aspects of their genetics and biology. Comparing with cold-water sponges that live for 11,000 years, for example, wouldn't be useful. This is why you need as many calibration points as possible when trying to get sensible results from broad-scale molecular clock analyses. For your work, you should probably try to find a mutation rate for a crustacean that is as close (phylogenetically and ecologically) as possible to yours.

Caveat: this isn't my area, and I'm haf-expecting a real specialist to disagree!

Hi Nozi,

Joseph is right! Mutation rates can be so different as higher as the difference between the taxonomic levels. Also, we have to bear in mind that mutation rates are driven by the balance of the replication system error rate and the efficiency of the repair systems per replication event (or generation). In other words, it's not linear for the whole genome or all species. As far you go from your taxonomic target more errors you'll get.

To solve that I do recommend you to use some fossil calibration or get as close as possible with the mutation rate estimation for your species.

To help you more, let us know what kind of data are you using for estimation? Is it one or more genes? Sometimes molecular clocks can trick you if you leave it strict with fewer points of calibration. Just imagine that if you have a lot of mutations in a low mutation rate, so you're telling to the software that those lineages took a long time to separate.

In addition, these relative mutation rates like something%/year doesn't look good for me. It's usually Xbp/(unit of time), this % thing sounds like 1,25% of the size of your gene, which can change everything (but I can be wrong about that).

You also should proceed some test for the molecular clock to see whether the region you are trying to use is a good region for a molecular clock or not.

You should also take in consideration that 53 mutations between lineages are a quite big number for just one gene, so with a constant mutation rate, you should take a bit of time (like some millions of years) to get there.

I think it does not solve your problem, but I hope it helps you a little bit.

Good luck.

Thank you so much for your answers. Unfortunately there are no fossil calibrations available for Tylos granulaus, the 1,25% that I used was estimated for a different group of isopods. To add more information, I used a COI gene. I'll also try to analyze my data with a different model to see if it makes any difference (Yule process).

Ok, I got it... What you could do is, if possible, to add sequences of those isopods which was used to calibrate such mutation rate. Then, if you have enough samples, use a coalescent method like Yule and look for patterns of tree branch diversification, as much more species you add to the tree more you'll be able to understand the process of diversification of that sequence in your subject group. Then, look out for how regular is the diversification of branches along the tree, it's qualitative but will give you an idea of how regular and reliable is the COI as molecular clock in your group. Also, you can use MEGA7 to test some hypothesis of a molecular clock. MEGA is easy to use and works with maximum likelihood to test the molecular clock hypothesis. By the way, you can also test the new algorithm from Tamura and Kumar (if I'm not wrong) they call it RelTime, it is in MEGA7 as well. Apparently, they did some comparisons and it has taken good results for trees with no regular mutation rates. I'm not sure if in such case the RelTime can give absolute divergence time, but it will give you a good idea of what is the variation of your gene among lineages. Take a look at "MEGA7: Molecular Evolutionary Genetics Analysis Version 7.0 for Bigger Datasets".

To deal with that, the best thing to do is follow those test I described you, but also use other regions more conservative. In addition, I've to notice you that in my experience with COI for vertebrates, 53 mutations is far than enough to differentiate even genus. However, because the problems I already report you, it should not be taken as a final conclusion. If you have sure that your lineages has some gene flow, so you're fine.

I hope you get luck.

Cheers.

There exists a model that accounts for the effects of body size and temperature on metabolic rate. By other words, it explains heterogeneity in rates of nucleotide substitution in different genes, taxa, and thermal environments. This model also suggests that there is indeed a single molecular clock, as originally proposed by Zuckerkandl and Pauling [Zuckerkandl, E. & Pauling, L. (1965) in Evolving Genes and Proteins, eds. Bryson, V. & Vogel, H. J. (Academic, New York), pp. 97–166], but that it ‘‘ticks’’ at a constant substitution rate per unit of mass-specific metabolic energy rather than per unit of time. This model therefore links energy flux and genetic change. By the way, the model allows to check the molecular dating done by the usual phylogenetic methods. Details at: Gillooly at al. 2005: The rate of DNA evolution: Effects of body size and temperature on the molecular clock. Proceedings of the National Academy of Sciences of the United States of America 102: 140-145.