[Biopython] EMBOSS and Python

[Peter]

Hi David,

I cc'd the mailing list again.

On Tue, Dec 22, 2009 at 4:02 PM, Lapointe, David
<David.Lapointe at umassmed.edu> wrote:
>
> Hi Peter,
>
> I have current version for both EMBOSS (6.1.0) and BioPython (1.53).

Do you have the original unpatched EMBOSS 6.1.0, or the latest
patched version, currently EMBOSS 6.1.0 patch 3? See:
ftp://emboss.open-bio.org/pub/EMBOSS/fixes/patches/README.patch

> I looked at the
> code for the unit tests (asis) and the problem might be there, as I could run the
> test fine by hand.

Could you clarify what you meant by "I could run the test fine by hand"?

> Shouldn't there be a '-' in front of asequence?
>
>    def test_water_file(self):
>        """water with the asis trick, output to a file."""
>        #Setup, try a mixture of keyword arguments and later additions:
>        cline = WaterCommandline(cmd=exes["water"],
>                                 gapopen="10", gapextend="0.5")
>        #Try using both human readable names, and the literal ones:
>        cline.set_parameter("asequence", "asis:ACCCGGGCGCGGT")
>        cline.set_parameter("-bsequence", "asis:ACCCGAGCGCGGT")
>
> David

Good question, but no. The test is confirming the set_parameter method
supports both these ways of setting the parameters. This is now a semi-
obsolete method - the preferred way would be to use bsequence in the
constructor arguments, or the bsequence property.

Interestingly the test output also indicates issues calling dnal (which
is nothing to do with EMBOSS), yet at least one other command line
tool test seem to be running OK (Clustalw).

Peter