[BiO BB] MrBayes - Memory overflow

Jose Maria Gonzalez Izarzugaza biopctgi at yahoo.es
Tue Feb 28 09:52:45 EST 2006


Many thanks.
Txema

Oscar Moya wrote:

>Ok, commands and computer discarded
>
>I have just seen that MrBayes has problems with 64 bit architechtures, don’t
>know if you have tried in 32 bits one. The following links speak about
>similar cases:
>
>http://sourceforge.net/mailarchive/forum.php?thread_id=9725904&forum_id=4511
>7
>
>http://www.rannala.org/phpBB2/search.php?search_author=GertW&
>
>
>A solution for some of them was to download the sourcecode from CVS at
>Sourceforge and compile it by themselves. 
>
>Hope that helps
>
>Óscar
>
>
>-----Mensaje original-----
>De: bio_bulletin_board-bounces+omoya=uib.es at bioinformatics.org
>[mailto:bio_bulletin_board-bounces+omoya=uib.es at bioinformatics.org] En
>nombre de Jose Maria Gonzalez Izarzugaza
>Enviado el: martes, 28 de febrero de 2006 12:53
>Para: The general forum at Bioinformatics.Org
>Asunto: Re: [BiO BB] MrBayes - Memory overflow
>
>Thanks a lot, Oscar.
>
>I've tried to run the consensus tree on different computers, with 
>different architechtures, and the results are the same in all of them: a 
>"nice" segmentation fault after reading the 2 tree files. The problem is 
>just with one specific dataset, the one with the most taxa. That's why i 
>say it migth be due to memory limitations.
>
>As i use a previously generated set of trees, the command block is as 
>simple as
>
>sumt filename=filewithALLtrees contype=allcompat burnin=1500 (where contype
>can also be skipped to the default half compat)
>
>----------------------------------------------------------------------------
>------------------------------------------------------------- 
>
>
>Oscar Moya wrote:
>
>  
>
>>Hi Txema,
>>
>>I am not sure about if your problem is due to memory needs. I am running
>>analyses of 70 sequences 1200 bases long in 512MB RAM computers.
>>
>>Have you tried to run it on another computer? It sounds evident, but
>>sometimes work.
>>
>>If problems keep, you could send the command block for checking (mcmc
>>settings, Lset, prset, etc...).
>>
>>Good luck!
>>
>>Óscar
>>
>>-----Mensaje original-----
>>De: bio_bulletin_board-bounces+omoya=uib.es at bioinformatics.org
>>[mailto:bio_bulletin_board-bounces+omoya=uib.es at bioinformatics.org] En
>>nombre de Jose Maria Gonzalez Izarzugaza
>>Enviado el: martes, 28 de febrero de 2006 11:52
>>Para: bio_bulletin_board at bioinformatics.org
>>Asunto: [BiO BB] MrBayes - Memory overflow
>>
>>Hello everyone.
>>
>>Does anybody know how i could limit the memory usage while creating a 
>>consensus tree with MrBayes? (even though i am using a 2GB RAM computer)
>>
>>The problem (segmentation fault) only appears when i'm trying to build a 
>>tree with more than 40 taxa, so the problem is not in the compilation, 
>>nor in the commands given.
>>
>>my mrbayes command is:
>>sumt filename=filewithALLtrees contype=allcompat burnin=1500
>>
>>Many thanks in advance.
>>
>>Regards,
>>Txema
>>
>>_______________________________________________
>>Bioinformatics.Org general forum  -  BiO_Bulletin_Board at bioinformatics.org
>>https://bioinformatics.org/mailman/listinfo/bio_bulletin_board
>>
>>
>>_______________________________________________
>>Bioinformatics.Org general forum  -  BiO_Bulletin_Board at bioinformatics.org
>>https://bioinformatics.org/mailman/listinfo/bio_bulletin_board
>>
>> 
>>
>>    
>>
>
>_______________________________________________
>Bioinformatics.Org general forum  -  BiO_Bulletin_Board at bioinformatics.org
>https://bioinformatics.org/mailman/listinfo/bio_bulletin_board
>
>
>_______________________________________________
>Bioinformatics.Org general forum  -  BiO_Bulletin_Board at bioinformatics.org
>https://bioinformatics.org/mailman/listinfo/bio_bulletin_board
>
>  
>




More information about the BBB mailing list