substitution rate vs. clock rate an & divergence dating

[rharp...@gmail.com]

Hi everyone,

I have a quick question regarding substitution rates vs. clock rates in BEAST2. Specifically, I am wanting to know exactly what these two parameters represent. Substitution rate parameter implies that this is the #mutations/persite/peryear. However, this link right below states that the average mutation rate (in persite/per year units) is actually specified under the 'clock.rate' parameter:

http://www.justinbagley.org/123/beast-and-the-beast-basics-molecular-clocks-and-how-to-input-rates-into-beast

For example, I am wanting to date the divergence between two sequences sampled from two species. For my dataset, I have estimates of the 'substitution rate' at 1.4% per million years. Using this blog, I get a 'substitution rate' as 7 x 10^-9.

(1.4/2)/1,000,000.

Now to fix this substitution rate, should I use this value for the 'clock.rate' parameter? Basically I am trying to understand the differences and the specific units that are used for the 'substitution rate' parameter (in the site model tab) and the 'clock.rate' parameter (in the clock tab). 

Additionally, following using the divergence dating tutorial here (http://beast2.org/tutorials/), I see that BEAST spits out the following parameters in the '.log file':

mutationRate.noncoding

mutationRate.1stpos

.......

clockRate

What is the differences between the 'mutationRate' and 'clockRate' parameters? Is the 'clockRate' parameter averaged across the partitions?

Thank you for the clarification, I am very interested in understanding exactly what these parameters represent (and the units provided in the .log file)

Thanks!

Rick

[Remco Bouckaert]

Hi Rick,

The terminology is a bit confusing, but the way BEAST uses them when using default settings is that substitution rates are interpreted as relative rates (relative to other partitions) and clock rates are absolute rates. So substitution rates are on average 1 when taking partition lengths in account, while clock rates are much lower than 1, e.g. 7 x 10^-9 if you set up your analysis in units of years (instead of say millions of years).

The times labelled "mutation rate” in the log are actually the substitution rates.

This may help setting up the analysis: http://beast2.org/2015/06/23/help-beast-acts-weird-or-how-to-set-up-rates/

Cheers,

Remco

--
You received this message because you are subscribed to the Google Groups "beast-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to beast-users...@googlegroups.com.
To post to this group, send email to beast...@googlegroups.com.
Visit this group at https://groups.google.com/group/beast-users.
For more options, visit https://groups.google.com/d/optout.