PacBio 'BaxH5Reader' object has no attribute '_ccsBasecallsGroup'

Question

[mjp]

Using smrtanalysis_2.3.0.140936 CLI I'm running into errors when following the example.t for smrtpipe_basemods.

For some other users, this type of error message meant wrong file type supplied in the input.fofn but I tried both: bas and bax and got to the same point. I also tried supplying both types at the same time but with no luck.

I do source the setup.sh environment before I execute smrtpipe.py if that is of any help.

Any insight on why this is happening would be appreciated.

Here is my P_Filter/filter_001of001.log:

[INFO] 2016-08-02 17:45:35,395 Successfully found 1 files in xml:input.xml
[INFO] 2016-08-02 17:45:35,980 smrtpipe.py running on localhost
[INFO] 2016-08-02 17:45:35,981 SMRT Analysis 2.3.0 / SMRTpipe 1.87.139483
[INFO] 2016-08-02 17:45:47,328 Process id 32153
[INFO] 2016-08-02 17:45:56,431 Starting task://Anonymous/P_Fetch/toFofn
[INFO] 2016-08-02 17:45:56,435 task P_Fetch.toFofn successfully completed (0.00 sec (0.00 min)). freed 1 slots, using 0 slots.
[INFO] 2016-08-02 17:45:56,435 Workflow Completion Status 1/38 in (            2%) tasks completed.
[INFO] 2016-08-02 17:46:01,456 Starting task://Anonymous/P_Fetch/adapterRpt
[INFO] 2016-08-02 17:46:01,459 Starting task://Anonymous/P_GenomicConsensus/writeContigList
[INFO] 2016-08-02 17:46:01,459 Starting task://Anonymous/P_ModificationDetection/writeContigList
[INFO] 2016-08-02 17:46:01,460 Starting task://Anonymous/P_Filter/filter.plsFofn.Scatter
[INFO] 2016-08-02 17:46:01,460 Starting task://Anonymous/P_Fetch/overviewRpt
[INFO] 2016-08-02 17:46:01,467 task P_GenomicConsensus.writeContigList successfully completed (0.00 sec (0.00 min)). freed 1 slots, using 4 slots.
[INFO] 2016-08-02 17:46:01,467 Workflow Completion Status 2/38 in (            5%) tasks completed.
[INFO] 2016-08-02 17:46:01,483 task P_ModificationDetection.writeContigList successfully completed (0.00 sec (0.00 min)). freed 1 slots, using 3 slots.
[INFO] 2016-08-02 17:46:01,484 Workflow Completion Status 3/38 in (            7%) tasks completed.
[INFO] 2016-08-02 17:46:01,499 task P_Filter.filter.plsFofn.Scatter successfully completed (0.00 sec (0.00 min)). freed 1 slots, using 2 slots.
[INFO] 2016-08-02 17:46:01,500 Workflow Completion Status 4/38 in (         . 10%) tasks completed.
[INFO] 2016-08-02 17:46:06,516 Starting task://Anonymous/P_Filter/filter_001of001
[INFO] 2016-08-02 17:46:10,237 task P_Fetch.overviewRpt successfully completed (5.00 sec (0.08 min)). freed 1 slots, using 3 slots.
[INFO] 2016-08-02 17:46:10,244 Workflow Completion Status 5/38 in (         . 13%) tasks completed.
[ERROR] 2016-08-02 17:46:10,245 *** Failed task task://Anonymous/P_Filter/filter_001of001
[INFO] 2016-08-02 17:46:48,405 Found 1 failed tasks.
[INFO] 2016-08-02 17:46:48,411 task filter_001of001 FAILED
Failed smrtpipe version v1.87.139483 with exit code -1 in 77.07 seconds (1.28 minutes)

[GenoMax]

Tagging: rhall

[mjp]

Just to add some more info to help in troubleshooting:

I checked the params.xml for possible misconfiguration but everything looked ok. I made sure the paths are setup correct, the reference points to the right file.

smrtpipe.log has one additional message that might help out:

[INFO] 2016-08-03 13:34:39,636 [SMRTpipe.SmrtPipeContext movieFiles 391] Found /home/username/data/sample/movie_s1_p0.bas.h5 (1318480 bytes)
[WARNING] 2016-08-03 13:34:39,640 [SMRTpipe.SmrtPipeMain _getBasVersions 419] Unable to correctly Parse the basH5 versions. Allowing job to proceed, but please fix the compatibility matrix under $SEYMOUR_HOME/common/etc. Error "unable to open object (Symbol table: Can't open object)"

I edited the filenames for simplicity of the report, they have original names for processing.

[GenoMax]

You can also contact PacBio tech support (support at pacb.com) to reach them directly for help.

[mjp]

That's what I just thought after I browsed through the SMRT Portal help pages although there is another address given: TechSupport at pacificbiosciences.com.

I think I will try both. Thanks for suggesting that. I'm kinda stuck here with no option to go further unless I receive some suggestion.

[GenoMax]

Both addresses probably reach the same team. Because of stuff like this it is safer to use SMRTportal GUI for analysis of data. Long as SMRTportal is correctly installed stuff works much better there than the command line.

[mjp]

I've read many successful stories about SMRT Portal. I, however, don't have access to GUI as I work on a small chunk of a larger cluster. Also my intention is to try to run things on a larger scale in the (hopefully) near future, so not sure how useful would the portal be at the end of the day.

I will see what the tech support comes back with.

[mjp]

I thought I will update this thread as I have managed to overcome this error message. It seems that it has been caused by the bas/bax files not being referenced in the right way.

It is strange though, that such cryptic message is displayed for relatively simple check that could be done even before the execution of the pipeline.

Unfortunately, I did find an official documentation in few places over the internet that suggested that both bas and bax files can be used as the right files in input.fofn. Using bas file format, although clearly bax is the correct one, didn't trigger an obvious warning which made the troubleshooting only harder and unnecessarily time consuming. Lesson learned.