Can't read log file in tracer

[Couvreur, Thomas]

Hello all,

A few months ago I posted a message about not getting appropriate ESS
values for my tMRCA even after 10 million generations. In fact I had to
run my analysis for at least 30 million generations for the ESS to be
above 100... . The problem is now that after a 50 million generation
analysis of 2 months, I can t get Tracer to read the .log file. It
starts to load but just at the end it blocks, and nothing happens. I
tried this on a mac and a PC several times and using the java version
and the executable one.
I wouldn t want to loose a 2 month analysis. The file is 600 mb, so it
is hard to open, on my PC can t at least.

If anybody has encountered such a problem already any advice on how I
could open this file in tracer would be much appreciated. Otherwise,
maybe there is some kind of bug (in tracer of in the making of the log
file) once passed the 40 million generations?
If someone wants to see the file, I could send via a ftp server such as
"sendit"or something...

Thanks again for all the great support we get for BEAST!

Thomas
---------------------------------------
Thomas L.P. COUVREUR Ph.D. student
Wageningen University, Biosystematics Group
National Herbarium of the Netherlands - Wageningen branch
Generaal Foulkesweg 37
6703 BL Wageningen
The Netherlands
email: Thomas....@wur.nl
http://www.bis.wur.nl/UK/Staff/Thomas+Couvreur/
Phone: +31 317 48 31 61
Fax: +31 317 48 49 17
Skype pseudo (free calling via internet): isolona21
-----------------------------------------------------------

[Andrew Rambaut]

Dear Thomas,

The problem is that the file is too big (Tracer loads it all into
memory). Generally, if you run BEAST for longer, you should reduce
the frequency at which it samples (we suggest aiming for 10,000
samples so in your case sampling every 5000 generations) - how often
did you sample?

Luckily you can use the program LogCombiner to down-sample to
something more reasonable. Run the program. Add your log file to the
table. (set burn in to 0 - you probably don't want to remove a burnin
until you can look at the file in Tracer). Give the new sampling
frequency in the box - this must be a multiple of the existing
sampling frequency. Then write the new log file.

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

[alexei....@gmail.com]

Hey Thomas,

If you are running a long analysis it may be better not to sample as
often as every 1000 states (the default). For most questions, there is
really no point in saving more than 1,000-10,000 samples in total,
especially if your ESSs are low. So if you are doing a 50M run you
should try to only sample ever 5,000-50,000 states... The easiest way
to fix you problem is to filter your file down to about 1/10 its
size... since your ESSs are only just above 100, you won't lose any
information by subsampling your file a bit. Tracer will probably choke
on files that are much greater than 100Mb.

Cheers
Alexei

> email: Thomas.Couvr...@wur.nlhttp://www.bis.wur.nl/UK/Staff/Thomas+Couvreur/

[Couvreur, Thomas]

Dear Andrew and Alexei,

Thanks for the help. That worked fine!
I was indeed sampling every 1000 generations. By resampling for every
5000 generations using LogCombiner I was able to open it in Tacer and my
ESSs are fine. After 50M they are now over 250.

Thanks for the great support!!

Thomas

[Mark Fenwick]

Hi

Has anyone loaded Beast onto a Vista system yet? I tried yesterday and
couldn't get it to download, any tips?

Cheers
M

Dear Andrew and Alexei,

Thomas

Dear Thomas,

Andrew

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Visit the Te Papa website http://www.tepapa.govt.nz
The email message together with the accompanying attachments may be
CONFIDENTIAL. If you have received this message in error, please notify
ma...@tepapa.govt.nz immediately and delete the original message. The
views expressed in this message are those of the individual sender,
except where the sender specifically states them to be views of Te Papa.
Te Papa employs strict virus checking measures and accepts no liability
for any loss caused either directly or indirectly by a virus arising
from the use of this message or any attached file.
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++