Blastn or Blastx?

[kalu]

Hi everyone,
I have just started annotating a transcriptome of a non-model organism (reads derived from a 454 sequencing) but would like to know what is the best algorithm to do it. I am wondering about blasting the assembled sequences using blastn to identify rRNA genes and genomic DNA and then blast the contigs left using blastx. Does it make sense to you?
Many thanks.

[nicedad]

Even though I do not know what do you mean with assembled sequences, almost what
usualy blasted are AllContigs or/and AllIsiotigs. Because these are the results of the assembling process.

cheers,

[kalu]

Dear nicedad,

Sorry if I did not make myself clear.. You are right, I meant to blast AllContigs. However my querry is related to which algorithm (blastn or blastx) is the best option to annotate them of a non-model organism.
Many thanks.

[Hobbe]

Did you check out Blast2go.org? I find it to be an excellent and very user friendly option for doing these kinds of annotations.

[A_Morozov]

Yeah, that makes sense. Though I think HMMs produce slightly better results when it comes to precisely defining position of RNA gene. For example, try this

http://www.cbs.dtu.dk/services/RNAmmer/

[nicedad]

In summary,
BLASTx searchs protein database using a translated nucleotide query whereas blastn for searching translated nucleotide database using a protein query.

best,

[kmcarr]

In summary,
BLASTx searchs protein database using a translated nucleotide query whereas blastn for searching translated nucleotide database using a protein query.

best,

Not quite. blastn searches a nucleotide database with a nucleotide query.

tblastn searches a translated nucleotide database with a protein query.

[zhidkov.ilia]

The right approach (most probably) will be to use blastx.
(taken from NCBI - tblastx takes a nucleotide query sequence, translates it in all six frames, and compares those translations to the database sequences dynamically translated in all six frames.)
Since you deal with non-model organism, the usage of all six frames in alignment is important.
Although blastx runs for very long time (compared to blastn), the obtained results are much more reliable and worth to wait for.

Ilia

[bso]

tBlastx for raw reads?

I am looking for a program that will align translated reads to a related genome (using tBlastx). Optimally, I'd like to visualize this against the translated genome. Maybe I should not try to do this with reads but instead assembled contigs? Any advice?

[grassgirl]

What is the advantage of using tblastx instead of blastx in this case? Wouldn't that return a lot of superfluous results since the correct reading frame is the actual protein or am I missing something?

[bso]

Blastx

Ok yes you're right, blastx-can you recommend a program that will allow me to visualize coordinate (going back to nucleotide) in the genome?

[grassgirl]

Hi bso,

I am not sure that I am clear on your objective (my original post regarding blastx was in response to kalu's inquiry...sorry if I didn't make that clear). You have your original read info, so are you asking how to get the nucleotide sequence from the blastx hits (as those will be amino acids)? If so, you can get that info by looking up the GenBank accessions on Entrez, click on the 'Gene' link (or narrow your search by entering Gene in the left dropdown menu) and scroll the the mRNA & protein section lower in the article on that gene, click on the mRNA accession number (should be the left one) and you will see the nucleotide sequence for that hit.

An automated script would be nice for running your hits through.