[Bioperl-l] Fate of Bio::Tools::PCRSimulation

Tue Mar 13 17:46:00 UTC 2012

Florent,

Getting a test fail on the amplicon branch, seems to be missing a file.

The only test consistently failing is t/Tools/AmpliconSearch.t (I'm 
seeing NCBI-related test issues that are also occurring on master). 
Using 'prove -lr t/Tools/AmpliconSearch.t':

[cjfields at pyrimidine bioperl-live (amplicons=)]$ prove -lr 
t/Tools/AmpliconSearch.t
t/Tools/AmpliconSearch.t .. 1/174
------------- EXCEPTION -------------
MSG: Could not open t/data/forward_primer.fa: No such file or directory
STACK Bio::Root::IO::_initialize_io Bio/Root/IO.pm:351
STACK Bio::SeqIO::_initialize Bio/SeqIO.pm:477
STACK Bio::SeqIO::fasta::_initialize Bio/SeqIO/fasta.pm:93
STACK Bio::SeqIO::new Bio/SeqIO.pm:358
STACK Bio::SeqIO::new Bio/SeqIO.pm:399
STACK Bio::Tools::AmpliconSearch::primer_file 
Bio/Tools/AmpliconSearch.pm:292
STACK Bio::Tools::AmpliconSearch::new Bio/Tools/AmpliconSearch.pm:151
STACK toplevel t/Tools/AmpliconSearch.t:126
-------------------------------------

# Looks like you planned 174 tests but ran 53.
# Looks like your test exited with 2 just after 53.
t/Tools/AmpliconSearch.t .. Dubious, test returned 2 (wstat 512, 0x200)
Failed 121/174 subtests

Test Summary Report
-------------------
t/Tools/AmpliconSearch.t (Wstat: 512 Tests: 53 Failed: 0)
   Non-zero exit status: 2
   Parse errors: Bad plan.  You planned 174 tests but ran 53.
Files=1, Tests=53,  0 wallclock secs ( 0.03 usr  0.01 sys +  0.18 cusr 
0.01 csys =  0.23 CPU)
Result: FAIL

chris

On 03/06/2012 11:08 PM, Florent Angly wrote:
>
> Yes, thanks Chris. If you want to start splitting a Bio-Tools
> distribution that would include Bio::Tools::AmpliconSearch, I am happy
> to help.
>
> In general, I am not specifically attached to the namespace, so if you
> guys prefer something different, just tell me.
>
> Note that AmpliconSearch uses a couple of new or re-worked objects,
> namely Bio::SeqFeature::Primer, Bio::SeqFeature::Amplicon and
> Bio::SeqFeature::SubSeq. Both Primer and Amplicons inherit from SubSeq,
> which inherits from Bio::SeqFeature::Generic. The main goal of SubSeq is
> to allow fetching a subsequence from the attached sequence or to
> explicitly adding a sequence that represents the feature (the Generic
> feature class does not support setting such a sequence, just getting the
> subsequence). I think it would be good to add this functionality to the
> Generic feature class, but I did not want to force things without asking
> everybody first if this seems like a good idea.
>
> Florent
>
>
> On 06/03/12 11:08, Fields, Christopher J wrote:
>> I'll check it out. Want me to post test results here (I have access to
>> a few systems to test on).
>>
>> chris
>>
>> On Mar 5, 2012, at 10:36 PM, Florent Angly wrote:
>>
>>> To all interested,
>>> the AmpliconSearch module is in a decent state. If you want to test
>>> it or improve it, head to
>>> https://github.com/bioperl/bioperl-live/blob/amplicons/Bio/Tools/AmpliconSearch.pm
>>>
>>> Regards,
>>> Florent
>>>
>>>
>>> On 01/03/12 12:42, Fields, Christopher J wrote:
>>>> Florent,
>>>>
>>>> Just want to add, my previous response isn't meant as an
>>>> admonishment, hope it didn't come across that way, but sometimes
>>>> email makes it hard to discern the difference. I simply meant to
>>>> demonstrate my opinion that I find releasing one's code is much
>>>> simpler (e.g. you can decide the rules and dictate when the code is
>>>> ready for release), and if we can make getting good code into user's
>>>> hands easier, more flexible, and more consistent I think that is
>>>> always a better path.
>>>>
>>>> chris
>>>>
>>>> On Feb 29, 2012, at 8:30 PM, Fields, Christopher J wrote:
>>>>
>>>>> There are a number of very good reasons to separate out common code
>>>>> and create new repos for new code. The problem about adding new
>>>>> code into core is it ties your code development to bioperl-live's
>>>>> release cycle and versioning. Also, what I (and others) would not
>>>>> like to see is any additional dependencies introduced, but a
>>>>> separate release allows you to (1) both add a dependency w/o
>>>>> affecting core, and (2) make it required, so no fiddling with
>>>>> checking for the module prior to running tests on it.
>>>>>
>>>>> As an example, I can easily see something like
>>>>> Bio::SearchIO::blastxml living on it's own since it has a set of
>>>>> outside dependencies.
>>>>>
>>>>> BTW, separation of modules into separate distributions (even single
>>>>> modules) based on functionality above and beyond that defined in a
>>>>> core is very common in the perl world. Beyond the obvious example
>>>>> of anything non-core in perl (all installable via CPAN), Moose,
>>>>> Dist::Zilla, Catalyst, Dancer, etc all have separately installable
>>>>> dists that layer additional functionality and have a separate
>>>>> maintenance path.
>>>>>
>>>>> chris
>>>>>
>>>>> On Mar 1, 2012, at 6:12 PM, Florent Angly wrote:
>>>>>
>>>>>> Thanks for everybody's feedback.
>>>>>>
>>>>>> I am looking at existing modules to hold template sequence,
>>>>>> amplicon sequence and primer information. There is the
>>>>>> Bio::SeqFeature::Primer and Bio::Seq::PrimedSeq. At the moment the
>>>>>> PrimedSeq object places Primer objects on the target sequence. I
>>>>>> have been looking at refreshing these modules (they are quite
>>>>>> old), add some sanity to them and make sure they are suitable for
>>>>>> a generic implementation of PCR (or amplicon search, which I find
>>>>>> a more suitable name since it is a far cry from simulating PCR
>>>>>> cycles, etc).
>>>>>>
>>>>>> I will make a remote branch today to make it easier for interested
>>>>>> parties to experiment and contribute.
>>>>>>
>>>>>> As you can see Chris, the amplicon search feature would use two
>>>>>> existing bioperl-live modules and only add one, tentatively in the
>>>>>> Bio::Tools::AmpliconSearch namespace. I am not convinced that this
>>>>>> warrants a separate distro.
>>>>>>
>>>>>> Florent
>>>>>>
>>>>>> On 01/03/12 01:23, Fields, Christopher J wrote:
>>>>>>> Seems like it was meant to be added at some point but was never
>>>>>>> committed. Definitely not in the github history for 1.3.x, this
>>>>>>> commit corresponds to the v1.3.4 tag:
>>>>>>>
>>>>>>> https://github.com/bioperl/bioperl-live/tree/0a67fa444eb19a70876017607f70ab72be38755a
>>>>>>>
>>>>>>>
>>>>>>> and it's not there.
>>>>>>>
>>>>>>> I agree with Roy, it would be nice to somehow make this a little
>>>>>>> more generic or pluggable on how it maps primers (maybe with a
>>>>>>> default pure perl method). I also think this shouldn't be bound
>>>>>>> to bioperl-live considering our current plans, it would best
>>>>>>> happen in a separate repo.
>>>>>>>
>>>>>>> chris
>>>>>>>
>>>>>>> On Feb 29, 2012, at 9:06 AM, Roy Chaudhuri wrote:
>>>>>>>
>>>>>>>> The code for Bio::Tools::PCRSimulation can be downloaded as part
>>>>>>>> of this archive:
>>>>>>>> http://www.salmonella.org/bioperl/primer3_v0.3.tgz
>>>>>>>>
>>>>>>>> (There's supposedly a more recent version here:
>>>>>>>> http://www.salmonella.org/bioperl/nucleotide_analyses.tgz
>>>>>>>> but that file seems to be truncated).
>>>>>>>>
>>>>>>>> I have no idea how much would be salvagable. It seems to just
>>>>>>>> use index to map the primers to the sequence, I guess it would
>>>>>>>> make more sense to at least give the option of something more
>>>>>>>> sophisticated like Primer3, BLAST or even a short read mapper.
>>>>>>>>
>>>>>>>> Cheers,
>>>>>>>> Roy.
>>>>>>>>
>>>>>>>>
>>>>>>>> On 27/02/2012 21:18, Fields, Christopher J wrote:
>>>>>>>>> On Feb 26, 2012, at 12:44 AM, Florent Angly wrote:
>>>>>>>>>
>>>>>>>>>> Hi all,
>>>>>>>>>>
>>>>>>>>>> I am interested in the Bio::Tools::PCRSimulation module.
>>>>>>>>>> Supposedly
>>>>>>>>>> it was added to Bioperl 0.3 and is also mentionned in the
>>>>>>>>>> Bio::PrimedSeq module. However, I cannot find in the current
>>>>>>>>>> Bioperl codebase. Any idea where it went?
>>>>>>>>> No idea; I can't find it anywhere in the code base either, and the
>>>>>>>>> github repo contains history going back to the original CVS repo.
>>>>>>>>> You can try contacting the author, possibly.
>>>>>>>>>
>>>>>>>>>> The reason I am asking is because I have some code to do
>>>>>>>>>> silico PCR
>>>>>>>>>> using regular expressions. I wanted to modularize my code more
>>>>>>>>>> and
>>>>>>>>>> make it into a module for Bioperl. Of course, if there is
>>>>>>>>>> something
>>>>>>>>>> similar in Bioperl already, I need to have a look at it. If there
>>>>>>>>>> is nothing similar, what namespace do you suggest to use?
>>>>>>>>>> Bio::Tools::AmpliconExtractor? Bio::Tools::AmpliconSearch?
>>>>>>>>>> Bio::Tools::InSilicoPCR?
>>>>>>>>>>
>>>>>>>>>> Thanks,
>>>>>>>>>>
>>>>>>>>>> Florent
>>>>>>>>> Maybe the last (InSilicoPCR).
>>>>>>>>>
>>>>>>>>> chris
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> _______________________________________________ Bioperl-l mailing
>>>>>>>>> list Bioperl-l at lists.open-bio.org
>>>>>>>>> http://lists.open-bio.org/mailman/listinfo/bioperl-l
>>>>> _______________________________________________
>>>>> Bioperl-l mailing list
>>>>> Bioperl-l at lists.open-bio.org
>>>>> http://lists.open-bio.org/mailman/listinfo/bioperl-l
>