[Biopython] blast to go annotation

[Peter Cock]

On Thu, Dec 6, 2012 at 11:59 AM, Fernando <fpiston at gmail.com> wrote:

> Yes,I'm. The script ignored the first set of results in the XML file.
> I had removed the line
> blast_record = blast_records.next()
>
> And it work Ok.
>

Good.

>> It would be worth reading the Blast2GO paper for some of the technical
>> issues and how to weight evidence in assigning GO terms based on
>> BLAST matches. Note Blast2GO has a command line variant called
>> "Blast2GO for pipelines" (b2g4pipe).
>>
>> Peter
>
> I know the Blast2GO. In fact, I started the GO annotation with that
> software, but I had many problems because it is very slow and crashes
> often.These problems make the annotation of many sequences with
> Blast2GO impossible.
> Furthermore, I also tried to use b2g4pipe in a cluster but the
> administrator told me also gives many problems. The administrator told
> me that the b2g4pipe has not been updated since their appearance and
> also requires an Internet connection. I think the free version of
> Blast2GO not be improved since they released the paid version.
>
> For this reason I decided to do it with my own code.

The future of Blast2GO with their paid version is troubling - they
never gave a clear answer on how the (money) free version is
licensed either.

However, b2g4pipe can be used without going online if you have
setup a local Blast2GO database - which is *much* faster than
connecting to their public database in Spain.

We're doing this locally as part of running b2g4pipe within Galaxy.
This includes a workaround for Blast2GO being unable to cope
with modern BLAST XML files (explained on their website with
links to other conversion scripts). You can find my code
in the Galaxy Tool Shed http://toolshed.g2.bx.psu.edu or
https://bitbucket.org/peterjc/galaxy-central/src/tools/tools/ncbi_blast_plus

Regards,

Peter