Error using pbdagcon

[JaLuxe]

I am currently testing you software and ran into a small problem:

I tried to perform the task manually from the terminal which resulted in the following output:

If i add spaces between the optional inputs and their respective values, pbdagcon works just fine as can be seen in the last picture.

is there anything i can do to get around this problem?

Best regards, JaLuxe

[JaLuxe]

changed it in my local "utils/buildConsensus.py" and it works like a charm.

edit: As i did not commit my changes, i will leave this issue open until somebody implements a fix in an update.

best regards, JaLuxe

[lch14forever]

Hi JaLuxe,

It seems you are using an older version of INC-Seq and a different version of PBDAGCON from me. I think the PBDAGCON version on our server was quite old (git commit #de1cf85) and I did not have problem with not including spaces for its arguments.

In a recent fix, I have included the compiled PBDAGCON binary used by us. In addition, I have implemented a function to avoid some non-terminating bug in PBDAGCON (I also added spaces between arguments and options).

def pbdagcon(m5, t):
    script_dir = os.path.dirname(os.path.realpath(__file__))
    cmd = ("%s/pbdagcon -t %d -c 1 -m 1  %s" % (script_dir, t, m5)).split()

    proc = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
    ## if in 5 sec, pbdagcon does not finish, trim 1 base and recursively run it
    poll_seconds = 0.25
    deadline = time.time() + 5
    while time.time() < deadline and proc.poll() == None:
        time.sleep(poll_seconds)

    if proc.poll() == None:
        proc.terminate()
        sys.stderr.write("Warning: PBDAGCON timeout! Trimming %d base(s).\n" %(t+1))

    stdout, stderr = proc.communicate()
    if proc.returncode != 0:
        stdout = pbdagcon(m5, t+1)
    return stdout

You can try to pull the latest changes, and this version uses only the PBDAGCON binary inside the "utils" folder.

Chenhao.

[JaLuxe]

Hi Chenhao,

First of all, thank your for replying so fast.

I have now freshly cloned the repository. Unfortunately, now another error occurs. The program call again is performed with the sample input in the data-folder. I only pasted the last few lines As the previous hundreds were identical.

File "/home/bio/Tools/INC-Seq/INC-Seq/utils/buildConsensus.py", line 220, in pbdagcon stdout = pbdagcon(m5, t+1) File "/home/bio/Tools/INC-Seq/INC-Seq/utils/buildConsensus.py", line 220, in pbdagcon stdout = pbdagcon(m5, t+1) File "/home/bio/Tools/INC-Seq/INC-Seq/utils/buildConsensus.py", line 220, in pbdagcon stdout = pbdagcon(m5, t+1) File "/home/bio/Tools/INC-Seq/INC-Seq/utils/buildConsensus.py", line 207, in pbdagcon proc = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE) File "/usr/lib/python2.7/subprocess.py", line 711, in __init__ errread, errwrite) File "/usr/lib/python2.7/subprocess.py", line 1207, in _execute_child if isinstance(args, types.StringTypes): RuntimeError: maximum recursion depth exceeded while calling a Python object

[lch14forever]

What if you replace the PBDAGCON binary with your version of PBDAGCON ($ mv utils/pbdagcon utils/_pbdagcon && cp `which pbdagcon` utils/)?

[JaLuxe]

good call!

Thank you a lot. It works :)

[lch14forever]

Could you let me know your PBDAGCON version?

The previous problem was because the older version somehow could go into infinite loop and the recursion was meant to fix that by trimming a few bases (although I don't understand how this fix it fully). And it seems the recursion went into infinite loop until stack overflow on your machine. I was not expecting the program to behave differently on different machines. I will leave this thread open, in case others face the same problem.

[JaLuxe]

I currently have no more access to the machine, but I downloaded it today via apt-get. Thus I think it should be the latest.

edit: formatting

[haansi]

Thanks for providing inc-seq and leaving this issue open :-)

For me the example call ./inc-seq.py -i data/inc_seq_test_read.fa -o consensus.fa resulted in issues with pbdagcon:

File "/../../INC-Seq/utils/buildConsensus.py", line 207, in pbdagcon proc = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE) File "/usr/lib/python2.7/subprocess.py", line 711, in __init__ errread, errwrite) File "/usr/lib/python2.7/subprocess.py", line 1207, in _execute_child if isinstance(args, types.StringTypes): RuntimeError: maximum recursion depth exceeded while calling a Python object

replacing with the compiled file with https://github.com/PacificBiosciences/pbdagcon (Latest commit 87f393f) solved the issue as well

[lch14forever]

Hi,

I have found a problem with the pbdagcon binary previously uploaded. I have uploaded the latest version of pbdagcon and tested it locally.

Chenhao.

[Szymonome]

Found that BLASTN switch generates some information at the end of the read name e.g. Basecall_2D_7/0_730 while POA does not. Why is it like that? Moreover, what exactly does the number "7" stands for? Initially I thought it may be number or molecules used for sequence correction however, fasta file from GRAPHMAP switch have reads with numbers 2, 3, 4 or 5 that shouldn't happen as minimum amount should be 6, f I'm right.

[lch14forever]

Hi Szymonome,

I have opened another issue for your question https://github.com/CSB5/INC-Seq/issues/6