[Biopython-dev] [Bug 3042] test_Mafft_tool fails

bugzilla-daemon at portal.open-bio.org bugzilla-daemon at portal.open-bio.org
Fri May 7 06:46:35 UTC 2010


http://bugzilla.open-bio.org/show_bug.cgi?id=3042





------- Comment #3 from mdehoon at ims.u-tokyo.ac.jp  2010-05-07 02:46 EST -------
(In reply to comment #2)
> (In reply to comment #0)
> > This is the error message I get:
> > 
> > ======================================================================
> > FAIL: Simple round-trip through app with infile.
> > ----------------------------------------------------------------------
> > Traceback (most recent call last):
> >   File "test_Mafft_tool.py", line 56, in test_Mafft_simple
> >     self.assert_("STEP     2 / 2 d" in stderr_string)
> > AssertionError
> 
> I have changed that to look for "Progressive alignment ..." instead
> which is present in both this MAFFT 5.x output and in MAFFT 6.x output.

This error has disappeared -- thanks!

> > > ======================================================================
> > FAIL: Round-trip with complex command line.
> > ----------------------------------------------------------------------
> > Traceback (most recent call last):
> >   File "test_Mafft_tool.py", line 126, in test_Mafft_with_complex_command_line
> >     self.assertEqual(return_code, 0)
> > AssertionError: 1 != 0
> 
> I've changed this to give the command line used to help debug when MAFFT
> returns an error code. Could you retest and report what MAFFT does for
> this particular command?

This is the output I am now getting:

======================================================================
FAIL: Round-trip with complex command line.
----------------------------------------------------------------------
Traceback (most recent call last):
  File "test_Mafft_tool.py", line 144, in test_Mafft_with_complex_command_line
    % (return_code, cmdline))
AssertionError: Got error code 1 back from:
mafft --localpair --weighti 4.2 --retree 5 --maxiterate 200 --nofft --op 2.04
--
ep 0.51 --lop 0.233 --lep 0.2 --reorder --treeout --nuc Fasta/f002

If I just run this mafft command directly, I get:

$ mafft --localpair --weighti 4.2 --retree 5 --maxiterate 200 --nofft --op 2.04
--ep 0.51 --lop 0.233 --lep 0.2 --reorder --treeout --nuc Fasta/f002
/usr/local/bin/mafft: line 184: [: --treeout: integer expression expected
Unknown option:  --treeout

MAFFT version 5.732 (2005/09/14)

    References:
      Katoh et al., 2002, NAR 30: 3059-3066
      Katoh et al., 2005, NAR 33: 511-518
      http://www.biophys.kyoto-u.ac.jp/~katoh/programs/align/mafft

    Options:
      --localpair :       All pairwise local alignment information is included
                          to the objective function. default: off
      --globalpair :      All pairwise global alignment information is included
                          to the objective function. default: off
      --op # :            Gap opening penalty (>0). default: 1.53
      --ep # :            Offset (>0, works like gap extension penalty).
                          default: 0.123
      --bl #, --jtt # :   Scoring matrix. default: BLOSUM62
                          Alternatives are BLOSUM (--bl) 30, 45, 62, 80,
                          or JTT (--jtt) # PAM.
      --nuc or --amino :  Sequence type. default: auto
      --retree # :        The number of tree building in progressive method
                          (see the paper for detail). default: 2
      --maxiterate # :    Maximum number of iterative refinement. default: 0
      --fft or --nofft:   FFT is enabled or disabled. default: enabled
      --memsave:          Memory saving mode (beta). default: off
      --clustalout:       Output: clustal format (not tested). default: fasta
      --reorder:          Outorder: aligned (not tested). default: input order
      --quiet :           Do not report progress.

    Input format: fasta format

    Typical usages:
      % mafft --maxiterate 1000 --localpair input > output
           L-INS-i (most accurate in many cases;
                    assumes there is only one alignable domain)

      % mafft --maxiterate 1000 --genafpair input > output
           E-INS-i (works even if there are many unalignable residues
                    between alignable domains)

      % mafft --maxiterate 1000 --globalpair input > output
           G-INS-i (suitable for globally alignable sequences)

      % mafft --maxiterate 1000 input > output
           FFT-NS-i (accurate and slow, iterative refinement method)

      % mafft --retree 2 input > output (DEFAULT; same as mafft input > output)
           FFT-NS-2 (rough and fast; progressive method)

      % mafft --retree 1 input > output
           FFT-NS-1 (very rough and very fast, applicable to >5,000 sequences;
                     progressive method with a rough guide tree)


> Also what is the output of "mafft --help" from MAFFT 5.732? That would be
> useful if we do have to make running the test conditional on the version of
> MAFFT.
> 

This is the output of "mafft --help":

$ mafft --help
Cannot open --help.

MAFFT version 5.732 (2005/09/14)

    References:
      Katoh et al., 2002, NAR 30: 3059-3066
      Katoh et al., 2005, NAR 33: 511-518
      http://www.biophys.kyoto-u.ac.jp/~katoh/programs/align/mafft

    Options:
      --localpair :       All pairwise local alignment information is included
                          to the objective function. default: off
      --globalpair :      All pairwise global alignment information is included
                          to the objective function. default: off
      --op # :            Gap opening penalty (>0). default: 1.53
      --ep # :            Offset (>0, works like gap extension penalty).
                          default: 0.123
      --bl #, --jtt # :   Scoring matrix. default: BLOSUM62
                          Alternatives are BLOSUM (--bl) 30, 45, 62, 80,
                          or JTT (--jtt) # PAM.
      --nuc or --amino :  Sequence type. default: auto
      --retree # :        The number of tree building in progressive method
                          (see the paper for detail). default: 2
      --maxiterate # :    Maximum number of iterative refinement. default: 0
      --fft or --nofft:   FFT is enabled or disabled. default: enabled
      --memsave:          Memory saving mode (beta). default: off
      --clustalout:       Output: clustal format (not tested). default: fasta
      --reorder:          Outorder: aligned (not tested). default: input order
      --quiet :           Do not report progress.

    Input format: fasta format

    Typical usages:
      % mafft --maxiterate 1000 --localpair input > output
           L-INS-i (most accurate in many cases;
                    assumes there is only one alignable domain)

      % mafft --maxiterate 1000 --genafpair input > output
           E-INS-i (works even if there are many unalignable residues
                    between alignable domains)

      % mafft --maxiterate 1000 --globalpair input > output
           G-INS-i (suitable for globally alignable sequences)

      % mafft --maxiterate 1000 input > output
           FFT-NS-i (accurate and slow, iterative refinement method)

      % mafft --retree 2 input > output (DEFAULT; same as mafft input > output)
           FFT-NS-2 (rough and fast; progressive method)

      % mafft --retree 1 input > output
           FFT-NS-1 (very rough and very fast, applicable to >5,000 sequences;
                     progressive method with a rough guide tree)



Thanks,
--Michiel.


-- 
Configure bugmail: http://bugzilla.open-bio.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.



More information about the Biopython-dev mailing list