psg_prepare_reference.py - java error

[Philip Jonsson]

Hello,

When I try to prepare my PSG models I get the following error:

Running on host compute-0-2.local

Time is Tue Nov 26 15:10:30 CST 2013

Directory is ...

Using 8 processors across nodes

Tue Nov 26 15:10:30 CST 2013

Using 8 cores

Reading transcripts from GTF-formatted standard input...

Read 57445 genes with 195584 transcripts and 1194206 exons

Processing chromosome chr1...

        Creating PSGs

Could not create the Java virtual machine.

Error: Command was terminated or failed:

java -cp /home/pkjonsso/local/psginfer-1.2.0/LearnPSG MakeFastaStdOut gencodev18$

Could not create the Java virtual machine.

Could not create the Java virtual machine.

Could not create the Java virtual machine.

Could not create the Java virtual machine.

Could not create the Java virtual machine.

Tue Nov 26 15:26:17 CST 2013

This is when I run it on our cluster. If I try to run it locally it seems to get past this step, but it's gonna take too long time. I have no idea what could at fault here. There is functional java on the cluster.

[Colin Dewey]

Hi Philip,

Could you give the version we just released (v1.2.1) a try?  This version passes an extra option to the Java virtual machine that should allow it to run more smoothly on multi-core machines.

Thanks,

Colin

--
You received this message because you are subscribed to the Google Groups "PSGInfer Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to psginfer-user...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Philip Jonsson]

Colin, 

I tried and get this (after all the java -cp calls):

bowtie-build -q gencodev18/references.fasta gencodev18/references

Exception in thread Thread-1 (most likely raised during interpreter shutdown)

Not sure what to make of it but it sounds like an error. The output directory seems to contain all the output files. 

[Colin Dewey]

Hi Philip,

Yes, I believe the output should be all fine in this case.  The error that you are receiving appears to be a currently unresolved issue with Python (http://bugs.python.org/issue9207) that appears every so often during shutdown of the Python interpreter.  It is unclear if there is anything we can do prevent this error message at this point.

Colin

[Philip Jonsson]

Alright Colin, thanks for the help.