Ok, thanks Peter for the detailed feedback, change recommendations, and your 
work detailing all the NCBI Blast fields and quirks over time!  

The columns 25 and 26 stem from my predecessor's version of the code which 
predates blast 2.28 I think (he had to create sallseqdescr from scratch).  I 
hadn't been paying any mind to how they fit in anywhere else in the last few 
years, but now that its being shared ... ! .  I'll take a close look at this, 
and nip the divergence in the bud so to speak - staying sync'd with your 
efforts is a plus.  I think the changes you mention will be fine for operations 
here.  (I'll have to see if sallseqdescr derived from XML hit titles is same as 
tabular salltitles).

Hmm, README.md is in github markdown variant so I'll see whats tripping up 
there.

Thanks,

Damion

Hsiao lab, BC Public Health Microbiology & Reference Laboratory, BC Centre for 
Disease Control
655 West 12th Avenue, Vancouver, British Columbia, V5Z 4R4 Canada
________________________________________
From: Peter Cock [p.j.a.c...@googlemail.com]
Sent: Tuesday, March 03, 2015 1:23 AM
To: Dooley, Damion; galaxy-...@lists.bx.psu.edu
Subject: Re: Galaxy tool shed alert for new repository named blast_reporting

Hi Damion,

I see your blast_reporting tool is now on the Main Tool Shed.
https://toolshed.g2.bx.psu.edu/view/damion/blast_reporting

I am a little confused about your tabular output options:

<option value="std" selected="True">Standard 12 columns</option>
<option value="std+seqs">Standard 12 columns + sequences</option>
<option value="ext">Extended 24 columns</option>
<option value="ext+">Extended 26 columns</option>
<option value="custom">Only field selections below</option>

In comparison my blastxml_to_tabular currently offers:

<option value="std" selected="True">Tabular (standard 12 columns)</option>
<option value="ext">Tabular (extended 25 columns)</option>
<option value="cols">Tabular (select columns to output)</option>

For the moment my pick-you-own columns choices are limited here
to the 25 columns (unlike the BLAST+ wrappers which offers more).
See: https://github.com/peterjc/galaxy_blast/tree/master/tools/ncbi_blast_plus

In ncbi_blast_plus the extended output used to be 24 columns,
my 25th column is the description (BLAST+ field salltitles, added
in BLAST+ 2.2.28).

https://github.com/peterjc/galaxy_blast/commit/0b161baa7f87d514c94fd1b3976af20bdc00c4c9
https://github.com/peterjc/galaxy_blast/commit/5234e32a277595044ec6d2a7002d93f9fb3011c7
http://blastedbio.blogspot.co.uk/2012/05/blast-tabular-missing-descriptions.html

Your in-tool help and the blast_reporting.py comments show
you've picked different extra columns:

Column    25 pcov          Percentage coverage
Column    26 sallseqdescr  All subject Seq-descr(s), separated by a ','

I'm guessing these names map to NCBI BLAST+ fields qcovs or
qcovhsp, and salltitles? Could you check and adopt the NCBI
names?

It is unfortunately the choice of fields has diverged.

Could you switch your columns 25 and 26 round in your "ext+"
output so that we agree on the first 25 columns? And expand
your "ext" output to add the descriptions and so match the 25
column extended output in ncbi_blast_plus too?

Thanks,

Peter

P.S. There is a problem in your markdown for the README.md
file, shown as a "Unexpected indentation." error when viewed on
the Tool Shed. https://toolshed.g2.bx.psu.edu/view/damion/blast_reporting


On Tue, Mar 3, 2015 at 1:46 AM,
<galaxy-no-re...@montana.galaxyproject.org> wrote:
>
> Sharable link:         
> https://toolshed.g2.bx.psu.edu/view/damion/blast_reporting
> Repository name:       blast_reporting
> Revision:              0:7db7ecc78ad6
> Change description:
> Uploaded
>
> Uploaded by:           damion
> Date content uploaded: 2015-03-02
>
___________________________________________________________
Please keep all replies on the list by using "reply all"
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:
  https://lists.galaxyproject.org/

To search Galaxy mailing lists use the unified search at:
  http://galaxyproject.org/search/mailinglists/

Reply via email to