TreeAnnotator

[Ainur Mukhatayeva]

Dear all,

I am trying to annotate my BEAST results in TreeAnnotator. For it, I am using .trees file and as burning states, I try 25.000.000, because my BEAST was running for 250mln runs. So, for TreeAnnotator it takes too much time, half of the day and I still can't see the end of it. Could you please help me with this issue? Should I decrease the number of burning states? As it was recommended to use for it the 10% from runs.

Thank you a lot!

Best wishes, Ainur

[Jordan Douglas]

Hi Ainur,

Tree annotator can be very slow when there are thousands of trees. I suggest thinning the posterior tree file using the beast/bin/logcombiner application. Use the -resample option to specifiy how much to thin eg. logcombiner -resample 10000 will include every 10,000 states

As for selecting the burnin, you should first open the log file in Tracer and see if the chains have converged, and what burnin is needed to achieve convergence. Sometimes it's more than 10%. The burnin % to achieve convergence in the log file should be the one you use for treeannotator or logcombiner

Hope this helps,

Jordan

[Ainur Mukhatayeva]

Hello Jordan

Thank you so much for your help! I found out that our computer was full of data and that is why the tree annotator was running for too long.

Now I did it in another computer, but I have got an error. 

Could you please let me know what is wrong there? I have attached the screen shot.

Thanks a lot for your help!

среда, 1 марта 2023 г. в 01:25:46 UTC+6, Jordan Douglas:

[Jordan Douglas]

Hi Ainur,

This problem is caused by running out of memory due to the input tree file being too large. There is a low memory option, which is slightly slower than the default. If you run treeannotator with "-lowMem" then it should be okay

BTW 25,000 trees is a lot of trees, and is probably more than you need. If the above suggestion does not work, or if the runtime is still too slow, I suggest further downsampling to perhaps 1000 - 10,000 trees using logcombiner.

Jordan 

[Ainur Mukhatayeva]

Hello Jordan,

Thank you so much for your help. I really appreciate it a lot.

I am not quite familiar with coding/software/computer science, and I am learning now. I attend the coding course. I started from Python, but I will learn others as well in the future.

Does the "-lowMem" function needs to be run as a code? 

I am actually using the TreeAnnotator as a program, so I could not find the "-lowMem" function there.

I will attach the screenshot of the tree annotator program that I use here.

l

Also, last time I checked my .log file in the tracer, I attach it here. It was saying "Burn in 25 000 000", that is why in tree annotator I put 25 000 000 in the "Specify the burning as a number of states: Burnin as states" line. I was running this file in the BEAST for 250 mln runs.   

Thank you!

Best wishes, 

Ainur

среда, 8 марта 2023 г. в 02:06:23 UTC+6, Jordan Douglas: