Node ages of up to 0. 08 for identical sequences??

[Inga]

Hi all,
I am currently trying to date when certain populations of one species split from each other by using a strict clock and a fixed clock rate (5% per MY). One of these populations suffered a recent bottleneck so all sequences I used in this analysis for this particular population are identical. However, the trees I have been getting show node ages of between 8,000 to 80,000 years between some of these individuals. Can someone help me to understand why this is? I have used a range of tree priors (Coalescent constant, Coalescent exponential, Bayesian Skyline and Yule (which I guess is generally ot appropriate for single species data??)), all with similar results.
Thank you very much!
Inga

[Alexei Drummond]

To get a handle on whether this is reasonable, ask yourself the following question: For a given substitution rate and sequence length, how much time could pass without seeing any substitutions at all.

If I assume 5% per million years, that is 0.05% per 10,000 years. So for a 500 nucleotide alignment I would expect 0.25 substitutions in 10,000 years (along a single branch), and the probability of 0 substitutions would be about p=0.77. For 100,000 years the probability of 0 substitutions would be 0.08. 

So assuming your alignment is about 500 nucleotides, your results don’t seem too crazy to me. If you want to be sure that the age of that cluster is younger you will need to sequence a longer fragment, and show that they are still identical over that longer fragment.

Cheers

Alexei 

--
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 http://groups.google.com/group/beast-users.
For more options, visit https://groups.google.com/d/optout.

[Inga]

Thanks Alexei, that makes it a lot clearer. I am just running an analysis on a longer fragment.
All the best
Inga