blast output to clustalw

[carolW]

Hi,
Which prorgams convert the tblastn output to clustal family program? How about blastp output?

Look forward to your reply,

Carol

[GenoMax]

"clustal" as in multiple sequence alignment family of programs?

Are you referring to getting "hits" from a blast search in fasta format so they can be fed into clustal?

[carolW]

yes.

the blast output contains the hits in effect but not in the fasta format. it's in the text format in which the hits are presented by NCBI reference sequence, description, score, identity and e-value

[GenoMax]

If you are doing your blast at NCBI then you can download fasta format sequences for the hits by using the download option on the results page (see attached screenshot).

If the blast is being done locally then you can extract the hits switching to tabular output for your blast and then retrieve fasta format sequences from the blast database using blast command line utilities as described here: http://seqanswers.com/forums/showthread.php?t=8805

[carolW]

OK thanks. fasta format for the output is not available in the download menu on the top of blast output result. That's why I thought that it's not available at all.

In addition to clustal, is there any known good soft for phylogeny? I know that there are phylip, treeview etc but don't know if any body is experience with this and can share his experience.

[GenoMax]

Take a look at MEGA, if you want a desktop based alternative.

[ssully]

mview

[carolW]

and to view the tree, in addition to treeview, which tools could be used? If the tree is saved in the fasta format, can clustal be used just for viewing?

What would you recomment to use as output format and viewing in terms of simplicity, less time and memory consuming, efficiency etc?

[carolW]

mview

I have a 450M fasta file of protein seq which is too large to upload in mview. What is the solution?

[GenoMax]

I don't think a web based solution is going to work with data of that size.

Is this an exploratory project where you are just interested in seeing a "tree" as to how the sequences are related or do you intend to do some serious phylogenetic analysis?

[carolW]

Take a look at MEGA, if you want a desktop based alternative.

Unfortunately, it doesn't seem that there is a linux version

[carolW]

I don't think a web based solution is going to work with data of that size.

Is this an exploratory project where you are just interested in seeing a "tree" as to how the sequences are related or do you intend to do some serious phylogenetic analysis?

I want to do some serious phylogentic analysis but I'm not sure of my soft choice yet. I'm just using an extract of my data but if this or other software will be my choice, I will have a bigger data file unless I can use other supported formats more compact than fasta. So I can consider the program running on unix but if I can optimize in terms of data format, I will be happy to change the input format

[carolW]

I don't know if anybody is experience with treeview but it crashed when I just uploaded a 5k phylip format file,

[carolW]

should I select complete fasta format or aligned fasta format for clustal for multiple or pairwise alignment and building tree?

the fasta format contains nucleotide sequence. it seems that there is no way to have fasta file with amino acid sequence? even if I do tblastn, why not?

If you are doing your blast at NCBI then you can download fasta format sequences for the hits by using the download option on the results page (see attached screenshot).

If the blast is being done locally then you can extract the hits switching to tabular output for your blast and then retrieve fasta format sequences from the blast database using blast command line utilities as described here: http://seqanswers.com/forums/showthread.php?t=8805

[GenoMax]

should I select complete fasta format or aligned fasta format for clustal for multiple or pairwise alignment and building tree?

the fasta format contains nucleotide sequence. it seems that there is no way to have fasta file with amino acid sequence? even if I do tblastn, why not?

Use complete fasta format for clustalW and related analysis.

Fasta format can include either nucleotide or protein sequence. If you are going to search for homologous sequences then starting your search with a protein sequence would be much better. Look at the delta-blast option at NCBI. It does seem to work much better then manually doing multiple psi-blast rounds.