[BioC] Automated blasting of short nucleotide sequences against each other

[Sean Davis]

Ken,

Actually, if you think about how blast (or blat or other alignment 
programs) works, you just need to blast the fasta against the blast 
database of the same sequences.  You will get output from blast that 
includes each sequence blasted against all others, with the obvious 
caveaut that not all sequences are going to align, so there will be 
some missing comparisons--no way around that.  Then, you just need to 
put them into some useful form--consider using bioperl if you have 
access to it.  Better yet, if these are sequences from the same 
organism, just use blat and the output is tab-delimited text which you 
can load directly into R.  If you use blat, you can just do:

blat db.fasta db.fasta outfile.psl

This should take just a few seconds on a modern machine, depending on 
the length of the sequences.

Sean

On Feb 18, 2005, at 2:41 PM, Ken Termiso wrote:

> Hi all,
>
> This may be slightly off-topic, but I'd like to be able to BLAST a 
> large set of about 500 nucleotide sequences against itself (i.e. 
> sequence #1 gets blasted against the other 499 sequences and so on, 
> for a total of 500x500 or 250,000 blasts), and one thing I 
> unbelievably cannot google on the net is a script to do it...rather 
> than writing one I was hoping that someone could point me to a link 
> for this...I found tons of scripts for doing it against a database, 
> but nothing with a matrix like I need to BLAST...
>
> My sequences are in plain text. I've got the standalone blast, but 
> just need a script...
>
> Presumably this would be very useful for analyzing pseudo-homologous 
> probe sequences..?..so maybe it isn't completely off-topic...
>
> Thanks in advance,
> Ken
>
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/bioconductor