HyPhy message board | |
http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl
HYPHY Package >> HyPhy bugs >> SBP bug http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl?num=1281045026 Message started by Matt on Aug 5th, 2010 at 2:50pm |
Title: SBP bug Post by Matt on Aug 5th, 2010 at 2:50pm
I am encountering an error when trying to run SBP analysis on my local machine. this happens when running both ./HYPHYMP (manually inputting options) and ./HYPHYMPI (using batchfile). This is using rev914, checked out today.
It is related to using the Model Option: Global w/variation+HM. The error (below) does not happen when not using this option. batchfile: Code (]inputOptions = {}; inputOptions ["00"):
the error: Code (]macmanes@macmanes:~/hyphy/HYPHY$ /home/macmanes/apps/openmpi-1.4.2/bin/mpirun -np 16 ./HYPHYMPI /home/macmanes/Desktop/matt_dqagenus.bf -------------------------------------------------------------------------- MPI_ABORT was invoked on rank 0 in communicator MPI_COMM_WORLD with errorcode 1. NOTE: invoking MPI_ABORT causes Open MPI to kill all MPI processes. You may or may not see output from other processes, depending on exactly when Open MPI kills them. -------------------------------------------------------------------------- There are 63 potential breakpoints. How many KH samples should be drawn per breakpoint ? How many rate classes should we have? Master node received an error: While attempting to construct category variable c: HMM is not a valid HMM-component model (square matrix of dimension 7) identifier in call to 'category' Call stack 3 : Category variable: {c,resp+1,catFreqMatrix,MEAN,(1-P)*GammaDist(_x_,alpha,alpha)*(_x_>0),(1-P)*CGammaDist(_x_,alpha,alpha)*(_x_>0)+P,0,1e25,(1-P)*CGammaDist(_x_,alpha+1,alpha)*(_x_>0),HMM} 2 : Select Template Model for {filteredData} 1 : ExecuteAFile from file DIRECTORY_SEPARATOR+"home"+DIRECTORY_SEPARATOR+"macmanes"+DIRECTORY_SEPARATOR+"hyphy"+DIRECTORY_SEPARATOR+"HYPHY"+DIRECTORY_SEPARATOR+"TemplateBatchFiles"+DIRECTORY_SEPARATOR+"SingleBreakpointRecomb.bf" using basepath /home/macmanes/Desktop/. reading input from inputOptions [macmanes:03600):
seems like a bug to me... |
Title: Re: SBP bug Post by Sergei on Aug 5th, 2010 at 2:52pm
Hi Matt,
Yeah - it is a bug (in the way how Global w/variation + HM is defined in the model file). I have not updated these model definitions in a long time and haven't used them myself. I bet ANY analysis that uses Global w/variation + HM codon model will die. I'll add this to my 'to-fix' list, thanks for the report. Sergei |
HyPhy message board » Powered by YaBB 2.5.2! YaBB Forum Software © 2000-2024. All Rights Reserved. |