I saw that tblastn is capable of doing that kind of search, but I want my position-specific scoring matrix (pssm) to be built from nucleotide, not protein, sequences.
Has anyone done something similar before?
Thanks!
You have a nucleotide PSSM and want to scan it against a database of nucleotide sequences?
Option 1: PoSSuM - excellent fast optimized PSSM scanner http://bibiserv.techfak.uni-bielefeld.de/possumsearch
Option 2: Instead of a PSSM, use the same alignments to build a nucleotide HMM and use NHMMER from HMMER 3.1-dev http://hmmer.janelia.org/
Option 3: Turn it around and use RPS-BLAST (rpsblastn from BLAST+) using your PSSM as the database and the sequences as your query.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
Hi Panos,
I have used Prophecy tools from Emboss package for something similar task !! You may have a look on that.
To add more , Prophecy create matrix based on 1. Gribskov 2. Henikoff 3. Frequency method too. I have used this long before during my master's thesis !! :) :)
Thanks a lot! I guess that after you create the pssm with Prophecy, you feed this to tblastn, right?
No !! First I did the alignments of sequences ( In my case we have transcription binding sites region) and then we created the matrix. .
So what did you do with the matrix? In my case, I want to blast it against a nucleotide database. Did you also do something like this?
I scanned sequences database with that Matrix !!
I scanned sequences database with that Matrix !!
I'm guessing you haven't used the matrix for searching a BLAST-formatted, nucleotide database, right? If you did, what program you used?
I used the Profit for scanning the Matrix with Sequences !! These sequences are 500 Upstream of IdeR Transcrition binding withing species of MTb !!