[rhall]

Also try running with the --debug option

[juassis]

wasn´t generated log, I can't even run the smrtpipe.py --help.
sometimes the pipeline functioned.

distribution:

Red Hat Enterprise Linux Server release 6.4 (Santiago).

thanks =)

[rhall]

Unfortunately this is due to the system - build system mismatch, and SMRT Analysis cannot be easily rebuilt. My only suggestion would be to run a virtual machine using something like virtual box installing a Ubuntu 10.04 system then installing SMRT Analysis on top of that.

[juassis]

rhall,
do you work for pacbio? It has a support contact?

It's impossible work in another machine

Many thanks for your help. =)

[rhall]

I work for PacBio, but not in tech support. If the installation is tied to a machine then go through your FAS for SMRT Analysis install support.

SMRT Analysis is not supported on Red Hat 6.4. You do not have to work on another machine, running virtual box on the Red Hat system will allow you to install a virtual Ubuntu 10.04 system for running SMRT Analysis.
For a discussion of the options, including using amazon AMI see:
http://seqanswers.com/forums/showthr...6421#post96421

[juassis]

unhh, ok!
I'll try the running virtual box!

thanks

[GenoMax]

I work for PacBio, but not in tech support. If the installation is tied to a machine then go through your FAS for SMRT Analysis install support.

That solves that mystery. You had very specific info that normal users do not (and we have had pacbio for a while).

Juassis: PacBio has a customer portal at: http://www.pacbioportal.com/. I am not sure if you need to have an instrument to get access. As rhall suggested you may need to go through your FAS.

[yaximik]

After browsing through the PacBio site I got impression that now the system is self-contained, that is both primary and secondary analysis are done locally. Is that correct, or raw data are still uploaded for primary analysis to PacBio's server and then send back for secondary analysis done locally? That would make it quite dependable on network reliability and the transfer rate, which varies a lot on my local network.

[rhall]

Primary analysis is done on the machine, the data is then transferred off the machine for secondary.

[yaximik]

Primary analysis is done on the machine, the data is then transferred off the machine for secondary.

I guess I was not clear. My understanding was that since beginning primary processing (base calling, etc.) was done somewhere at PacBio uusing their proprietary software. Correct me if I am wrong.

Now it seem all done localy, I mean on the instrument (primary) and off instrument, on some separate datarig (secondary).

It is mentioned that a blade cluster is the part of the instrument, I guess inside their big box, intended for primary processing. Then a separate big cluster is needed for secondary analysis, which is not the part of the package, correct?

I sent a info request to PacBio, hope their sales people get back to me and explain all that.

[rhall]

Primary analysis, base calling, has always been done on the blade cluster, which is part of the instrument.
A separate cluster is needed for secondary analysis tasks, that is not part of the package that PacBio sells, the size of this depends on what you want to do and can range from a high end workstation to a 100 node cluster.

[GenoMax]

Then a separate big cluster is needed for secondary analysis, which is not the part of the package, correct?

Rhall has already explained the basics. You can download the secondary analysis software (both command line/web front end) here: http://pacbiodevnet.com/.

Note: Secondary analysis software is only supported on CentOS and Ubuntu (note the specific versions there in). Stepping outside those will likely prove an exercise in futility (i.e. do not try it, in most cases it does not work). If you are running a OS version that is not one of the two supported then consider using a virtual server.

[rhall]

For anyone still having issues with SMRT Analysis installations, I'm experimenting with a simpler, somewhat limited, install method using Vagrant and VirtualBox.

https://github.com/PacificBiosciences/Bioinformatics-Training/wiki/SMRT-Analysis-2.0-Virtual-Machine-Install-(Experimental)

[juassis]

[QUOTE=rhall;98498]I work for PacBio, but not in tech support. If the installation is tied to a machine then go through your FAS for SMRT Analysis install support.

Cheers for your answer, it was very helpful.
But, I don't understand the error message generated, with solid datas.

Solid - library mate pair - reads = 2.159.990.294 - coverage 37x (151 GB)

error log message as below:

[ERROR] 2013-05-20 14:24:19,051 [pbpy.smrtpipe.engine.SmrtPipeWorkflow execute 602] task://Anonymous/P_Filter/filter_001of001 Failed
Traceback (most recent call last):
File "/opt/smrtanalysis-1.4.0/analysis/lib/python2.7/pbpy-0.1-py2.7.egg/pbpy/smrtpipe/engine/SmrtPipeWorkflow.py", line 600, in execute
raise WorkflowError(task.error)
WorkflowError: task://Anonymous/P_Filter/filter_001of001 Failed

[ERROR] 2013-05-20 14:24:21,218 [pbpy.smrtpipe.SmrtPipeMain run 648] SmrtExit task://Anonymous/P_Filter/filter_001of001 Failed
Traceback (most recent call last):
File "/opt/smrtanalysis-1.4.0/analysis/lib/python2.7/pbpy-0.1-py2.7.egg/pbpy/smrtpipe/SmrtPipeMain.py", line 608, in run
self._runTasks(pModules)
File "/opt/smrtanalysis-1.4.0/analysis/lib/python2.7/pbpy-0.1-py2.7.egg/pbpy/smrtpipe/SmrtPipeMain.py", line 281, in _runTasks
workflow.execute()
File "/opt/smrtanalysis-1.4.0/analysis/lib/python2.7/pbpy-0.1-py2.7.egg/pbpy/smrtpipe/engine/SmrtPipeWorkflow.py", line 605, in execute
raise SmrtExit(str(e))
SmrtExit: SmrtExit task://Anonymous/P_Filter/filter_001of001 Failed
[ERROR] 2013-05-20 14:24:21,238 [pbpy.smrtpipe.SmrtPipeMain exit 760] SmrtExit task://Anonymous/P_Filter/filter_001of001 Failed

---

thanks

[rhall]

Correct me if I'm wrong, are you trying to use SMRT Analysis with Solid sequence data? While some parts of SMRT Analysis may be able to handle Solid data, the workflow is really designed for PacBio data.
Trying to run the command in ./workflow/P_Filter/filter_001of001.sh should give you a more descriptive errror.