Extraction step not working

[ambarrio]

Hi!

After running support, I am trying to run the extraction step and it is not functioning for me. I was wondering what are others experiences with this step. Or if I could get any help from Adam. I am using PBSuite_14.7.14.

Let's say that I have the following simple graph - saved in the .gml file created after executing "support":

graph [
  node [
    id 0
    label "ref0001860e3"
    extenders "m140530_040015_42203_c100653532550000001823125010071481_s1_p0/139728/0_13347:m140524_082726_42203_c100657162550000001823122510071413_s1_p0/50941/15946_18572"
  ]
  node [
    id 4006
    label "ref0000288e5"
    extenders "m140524_082726_42203_c100657162550000001823122510071413_s1_p0/33477/2734_5148"
  ]
  node [
    id 4048
    label "ref0001973e3"
    extenders ""
  ]
  edge [
    source 4006
    target 4048
    evidence "m140529_155547_42203_c100653522550000001823125010071497_s1_p0/39343/0_14955"
  ]
]

After installing pyparsing-2.0.2 -which is not specified as required in the documentation- I start running "extraction". It dies with the following trace:

2014-09-09 18:14:02,502 [INFO] Loaded 60889 Reads
2014-09-09 18:14:03,584 [INFO] Parsed 9831 Reads
Traceback (most recent call last):
  File "~/glob/local/PBSuite/bin/Extraction.py", line 344, in <module>
    me.run()
  File "~/local/PBSuite/bin/Extraction.py", line 317, in run
    self.__extractReads__()
  File "~/local/PBSuite/bin/Extraction.py", line 203, in __extractReads__
    self.flushQueue(outputQueue)
  File "~/local/PBSuite/bin/Extraction.py", line 236, in flushQueue
    outFile = self.openGapOut(gapName)
  File "~/local/PBSuite/bin/Extraction.py", line 252, in openGapOut
    basedir = os.path.join(self.supportFolder, gapName.replace('/','.'))
AttributeError: 'int' object has no attribute 'replace'

gapName is not really a str but an int. It dies when calling openGapOut with gapName. gapName comes from calling self.flushQueue(outputQueue). If I log within the function:

       for gapName in outputQueue:
            logging.info("GapName in flushQueue: " + 
                str(gapName) )

            #pp.pprint(self.gapOutputs[gapName])
            pp.pprint(gapName)

it'd log to the .err file:

2014-09-09 18:14:03,585 [INFO] GapName in flushQueue: 0
2014-09-09 18:14:03,585 [INFO] GapName in openGapOut: 0

and 0 to the .out one with the pprint instruction.

As outputQueue is a :

defaultdict(<type 'list'>, {0: ['@m140530_040015_42203_c100653532550000001823125010071481_s1_p0/139728/0_13347\nAAAAGAGAGAGATTTCGGCG... and then it will come the qualities and so on...
]

I wonder what Jelly is trying to do in the replace instruction. As I think it is trying to replace the / with . in the identifiers of the PB read (@m140530_040015_42203_c100653532550000001823125010071481_s1_p0/139728/0_13347), or?

Then, if I use str() to qualify "0" as a string, it can continue, but it will die somewhere else afterwards:

Traceback (most recent call last):
  File "~/local/PBSuite/bin/Extraction.py", line 344, in <module>
    me.run()
  File "~/local/PBSuite/bin/Extraction.py", line 317, in run
    self.__extractReads__()
  File "~/local/PBSuite/bin/Extraction.py", line 203, in __extractReads__
    self.flushQueue(outputQueue)
  File "~/local/PBSuite/bin/Extraction.py", line 236, in flushQueue
    outFile = self.openGapOut(gapName)
  File "~/local/PBSuite/bin/Extraction.py", line 266, in openGapOut
    sequence = self.reference[flankName[:10]]
KeyError: '0'

Hope you can help me to sort out this problem. Very much looking forward to your feedback,

álvaro

[ambarrio]

Hi!

Just for the records. I was running the wrong version. Support does not work on v14.7.14 but it works on the most recent version, 14.9.9.

Problem solved.

álvaro