[Bioperl-l] A question about iBio::Index: and its correct use

[jluis.lavin at unavarra.es]

Thank you for the info Florent!
I´ll try to read al the information on the link you provided and try to
figure out how to make it work and if it is worthy for me, I mean, I work
with several sequence files that come from multiple databases (JGI, BROAD,
Genolevures or NCBI). Protein IDs from each of those databases is
different from NCBI. Maybe it could be easier to write a script that
allows me to enter a fasta file with all the protein models of a single
organism, parse it and then extract the sequences of a given list (using
the "ID style" of the particular database) than creating a BLAST index for
each organism I need to work with...Did I explain the issue correctly?
Anyway, since I don´t know anything about this tool Hans and you provided
me, I can easily be wrong...
Thank you for showing me the local BLAST Index tool, I´ll read the
documentation carefully and study all its possibilities.

Best wishes

JL


El Jue, 5 de Noviembre de 2009, 19:00, Florent Angly escribió:
> Hans-Rudolf was talking about a way to retrieve sequences from a BLAST
> database. If you use BLAST locally, then your database is local too.
> More info here:
> http://www.ncbi.nlm.nih.gov/staff/tao/URLAPI/formatdb_fastacmd.html
> Florent
>
>
> jluis.lavin at unavarra.es wrote:
>> Thanks a lot for your help Hans,
>> It's a little bit to hard to understand and turn into script this
>> awesome
>> information you've just given me...I hope I can use it in a near future
>> anyway ;)
>> The issue here is that the sequences I,m indexing are not generated by
>> the
>> NCBI nor stored there...although I belive you´re just refering to the
>> tool
>> itself and not to a retrieval from the NCBI.
>>
>> Thanks again you´re all great giving advice to newbies like me ;)
>>
>> Best wishes to you all
>>
>>
>> El Jue, 5 de Noviembre de 2009, 17:02, Hotz, Hans-Rudolf escribió:
>>
>>>
>>> Jluis
>>>
>>>
>>>> -Then I´ll beg you to take a look at my scripts, because I don´t seem
>>>> to
>>>> catch the bug...
>>>>
>>> you haven't attached/included any scripts, have you?
>>>
>>>
>>> Anyway, have you considered using BLAST indices (created with the
>>> additional
>>> flag "-o") together with the tool 'fastacmd' (which also included in
>>> the
>>> NCBI blast binaries) as a simple (and very fast) alternative for
>>> fetching
>>> sequences.
>>>
>>>
>>> Regards, Hans
>>>
>>>
>>>
>>>
>>
>>
>>
>
>


-- 
Dr. José Luis Lavín Trueba

Dpto. de Producción Agraria
Grupo de Genética y Microbiología
Universidad Pública de Navarra
31006 Pamplona
Navarra
SPAIN