Hi,
I've just completed a tutorial on divergence time estimation using BEAST (http://treethinkers.org/wp-content/uploads/2014/08/DivTime_BEAST2_tutorial_2014.pdf), but I have questions relating to my own data.
The above tutorial is based on taxa for which there are known calibration points. For the taxonomic group my Honours project data relates to there is no such calibration data. Topology and branch lengths in my preliminary ML tree tell me that I'm dealing with divergent clades and divergent species within clades; I think the Relaxed Random Local Clock model and the Birth & Death tree prior will best fit my data, my questions are:
1) In the absence of calibration data can BEAST estimate divergence times?, or will analyses result in divergence rate estimates?
2) Which priors or other important information do I need to specify in BEAUti while implementing the relaxed clock model and birth and death tree prior?
Any advice would be appreciated,
John
For the question 1, YES, BEAST estimates divergence times "natively". In other words, if you didn't input absolute time calibration points, you will retrieve relative divergence ages (you will know that the species X is 2x older than species Y, for example).
BUT, if you have some information about the substitution rates in your taxa for the data you use, you can add it to your model of evolution to retrieve absolute ages for your clades. I think is not an ideal solution, as the substitution rate can vary extensively, even in closely related species, but it will be better than nothing.
Finally, do you really need to obtain absolute ages ?
Hi Damien,
Thanks for the reply; since I first posted it has been gently pointed out to me that in the BEAST divergence time tutorial I mentioned above, it does actually state that 'substitution rates' can be used as calibration information (note to self: must concentrate harder when reading tutorials).
From the literature I have substitution rates per site (patristic distance) between genera and species within genera for a few of the taxa In my phylogeny. As you point out in your post substitution rates can vary within and between clades, and the rates I've obtained from the literature do vary (0.075 up to 0.394). To answer your question yes I'm hoping to end up with an estimation of divergence time for three lineages within one particular clade in my tree.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Programming Languages