Whats the difference between the *.log file created and the *.tree file created ?

[Anuj Guruacharya]

Hello All,

I am new to BEAST. Just began using it two days ago. 

FIrst question,

I would like to know Whats the difference between the *.log file created and the *.tree file created ? 

The tracer, log analyzer, and the log annotator all these programs seem to use only the log file. What use is the *.tree file then ?

Second question,

How can I input my own tree into Beast to do diversification rate analysis ?

Thanks,

Anuj

[Tatiana Aghova]

Hi Anuj,

congratulation for start with BEAST :)

1) log file is very important, you have there all information about the run. You can open it in Tracer and see, if the run is OK or not (you look on statistic and trace)

in the *.tree file you find all trees (if you have default, usually 10 000) you can visualise it all using program Densitree, or you can run TreeAnotator and you get one tree.

2) I am not sure, if I understand your second question. If you would like run molecular clock it is not necessary to use your own tree. BEAST is smart and can made it.

I hope I help you a little bit (or probably someone experienced will help you)

Good luck

Tatiana

Dňa štvrtok, 12. februára 2015 21:53:56 UTC+1 Anuj Guruacharya napísal(-a):

[Santiago Sánchez]

Just to add a little bit more, both ".log" and ".trees" are log files which contain the sampled states from the MCMC. For instance, if you run 10x6 generations and you sample every 1000 states, both your ".log" and ".trees" files will contain 10,000 samples (or states). The difference between both files lies in that one (.log) is in a table format and includes the sampled states of each and every parameter in your model (or suit of models, e.g. substitution model, molecular clock model, transition rates from discrete CTMC, calibrations, etc.), and their associated likelihood scores, including the prior and posterior. The ".trees" file includes essentially the same information for every sampled state, but in the form of annotated ultrametric trees. Regularly, you would use Tracer to visualize your ".log" file, and determine if your analysis reached convergence (e.g. by comparing multiple runs) and if there was sufficient mixing in the model parameters throughout the MCMC (by looking at the ESS values). If your analysis passed the test, regularly you would summarize the information in your posterior trees (.trees) into a single, most informative, maximum-clade-credibility (MCC) tree. Then you can visualize your results by opening your MCC tree in FigTree, for instance, to look at the topology, the divergence times, the level of rate variation among lineages, or the evolution of a discrete or continuous trait (and for each, their level of uncertainty, e.g. 95% HPD intervals).

On your second question, although BEAST can estimate (for tree building and parametric purposes) birth and death rates (i.e. speciation and extinction rates), normally you would do diversification rate analyses with other, more specialized, software, using your MCC tree or maybe a subsample of your posterior distribution of trees. Some of these programs or packages (e.g. BAMM, DDD, PBD, Diversitree, Geiger, phytools) can perform diversification rate analyses, but vary in the model implementation.

Cheers,

Santiago

--
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.

--

Santiago Sánchez-Ramírez

Department of Ecology and Evolutionary Biology, University of Toronto

Department of Natural History (Mycology), Royal Ontario Museum

100 Queen's Park

Toronto, ON

M5S 2C6

Canada

Other email: santiago...@mail.utoronto.ca

Tel. 416-586-8025

Website: https://sites.google.com/site/santiagosnchezrmirez/