Can Tracer work without GUI?

[Zheng Hou]

Hi,

Can Tracer run without the use of GUI, only from commod line? Since there are lots of independent runs to trace the statistics, using Tracer from the GUI seems impossible. Any help? Thank you!

Best,

Aroma

--
School of Life Sciences ,  Fudan University
220# Handan Road, Shanghai(200433)P.R.China.

[Andrew Rambaut]

There is a command-line program called LogAnalyser in the BEAST package that can analyse log files.

Best,
Andrew

> --
> You received this message because you are subscribed to the Google Groups "beast-users" group.
> To post to this group, send email to beast...@googlegroups.com.
> To unsubscribe from this group, send email to beast-users...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/beast-users?hl=en.

[Zheng Hou]

Dear Andrew, 

I'm now tring to use loganalyser in BEAST1.7.0,  but encounter the following the program: 1. the program has been running for more than two days while the output file is only 6kb in size; 2. If I want to generate the bsp plot of linear change of population size, the tree file should be analysed simutaneously but I donot know how to do this by loganalyser. If loganalyser can do this, could you please tell me how to use it? If not, could you please provide me the alogrithm used in Tracer? 

Best,

Zheng

2012/4/27 Andrew Rambaut <ram...@gmail.com>

[Andrew Rambaut]

Dear Zheng,

LogAnalyser doesn't do skyline reconstructions. There is a small command line app to do so but it was written specifically for someone who wanted to analyse the output of some simulations. However, it shows the basic algorithm. The source code is here:

http://code.google.com/p/beast-mcmc/source/browse/trunk/src/dr/app/tools/SkylineReconstructor.java

Hope this helps,
Andrew

___________________________________________________________________
Andrew Rambaut
Institute of Evolutionary Biology University of Edinburgh
Ashworth Laboratories Edinburgh EH9 3JT
EMAIL - a.ra...@ed.ac.uk TEL - +44 131 6508624





--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

[Zheng Hou]

Dear Andrew,

Thank you for you kindly help and the link below does help me a lot. But there is still details I'm not clear since the program depends on several other packages that I failed to search in the same website. They are listed as the following: 

import dr.inference.trace.LogFileTraces;
import dr.inference.trace.TraceDistribution;
import dr.inference.trace.TraceException;
import dr.inference.trace.TraceList;
import dr.stats.Variate;
import jebl.evolution.coalescent.IntervalList;
import jebl.evolution.coalescent.Intervals;
import jebl.evolution.io.ImportException;
import jebl.evolution.io.NewickImporter;
import jebl.evolution.io.NexusImporter;
import jebl.evolution.io.TreeImporter;
import jebl.evolution.trees.RootedTree;

 
I wrote a program which has the same basic algorithm with the one in the link you supplied but cannot get the same results as in Tracer.  Also, I want to make comparions with other methods. So, I hope to make the details clear. Could you please give me further help? Thank you very much!

Best,

Zheng

2012/5/28 Andrew Rambaut <a.ra...@ed.ac.uk>

[Zheng Hou]

Dear Andrew,

   Where can I get the source codes of Tracer? There are a lot of simulation output files from BEAST, so it is not possible to use Tracer. We have tried to write a program to construct skyline with the algorithm showed in the link you supplied. However, we can not get the same result as Tracer does: for each MCMC sampling state, straight lines are drawn between each neighboring two skyline.popsize* points, and if a time point that we are interested in is in the range of a certain line, then the corresponding point on the line is considered as one sample from the time point that are under interest. Two of us have written programs independently and we got the same result which is quite different from the one from Tracer. And we consider this may be caused by different methods of dealing with the *.(time).trees files, since we take the mutation rate in each branch as missing uninteresting data and just ignore them, only taking considerations of branch lengths. And may be reading the source codes of Tracer is the most direct way for us. Could you please give us some help?

Best,

Zheng

2012/5/28 Andrew Rambaut <a.ra...@ed.ac.uk>

Dear Zheng,