FASTA AND BLAST

hi, are there fasta and blast functions in matlab?

thanks

Answer 1

[Walter Roberson]

http://www.mathworks.com/help/toolbox/bioinfo/ref/blastlocal.html

http://www.mathworks.com/help/toolbox/bioinfo/ref/fastaread.html

http://www.mathworks.com/help/toolbox/bioinfo/ref/seqprofile.html

http://www.mathworks.com/help/toolbox/bioinfo/ref/multialign.html

[Jonathan]

Nice.

[huda nawaf]

thanks

but i did not find the heuristic software BLAST for in matlab.

can help me to find it if their .

[Walter Roberson]

Well, that is odd, considering that the blast* functions in that toolbox specifically include the reference,

Altschul, S.F., Gish, W., Miller, W., Myers, E.W., and Lipman, D.J. (1990). Basic local alignment search tool. J. Mol. Biol. 215, 403–410.

Answer 2

[Lucio Cetto]

None of these alignment algorithms are included in Matlab.

Do not mix the "fasta" alignment algorithm with the "fasta" file format. Matlab provides fastaread and fastawrite for I/O FASTA formatted files.

In the case of the "blast" algorithm Matlab does provides a wrapper that allows you to pass the information to a locally installed blast (blastlocal), and provides a read function (blastlocalread) that will let you put the results of the alignments back in Matlab, but you will need to install the blast algorithm (provided by NCBI) by yourself. Also Matlab provides a function that allows you to submit queries to the NCBI's blast server, balstncbi (i.e. aligning against their databases). The function blastread will read the web data back into Matlab.

[huda nawaf]

i know there is fasta file format and fasta alg.

my query now is:

can i use the blast alg. that provided by NCBI in case of alignment non biological seq.?

thanks

[Lucio Cetto]

This is an interesting question that I have come accross several times.

Matlab allows you to run an alignment (using a Needleman&Wunshtype algorithm) with symbols other than aminoacids or nucleotides, for intance look at this <http://www.mathworks.com/matlabcentral/newsreader/view_thread/281532#743065 thread>. Note that this is undocummented.

Also, if you want to align instead numeric values (or even vectors) you can look at the |samplealign| function. It also uses a dynamic programming algorithm. Run the examples in the m-help, this is a function hard to understand but very handy once you understand how it works.

Note that in both of the solutions I give, none of them use an heuristic approach (such as in the BLAST or FASTA algorithms).

HTH

[huda nawaf]

thanks Lucio

I did not see this last comment , walter draw my attention to this comment.

i will see that link and that function

Answer 3

[Jonathan]

The following commands and output says there are not such functions.

>> help fasta

fasta not found.

Use the Help browser Search tab to search the documentation, or
type "help help" for help command options, such as help for methods.

>> help blast

blast not found.

Use the Help browser Search tab to search the documentation, or
type "help help" for help command options, such as help for methods.

If you can tell us the purpose of the functions, perhaps we can answer if there are equivalent functions in Matlab.

[huda nawaf]

It is heuristic algorithms of local sequence alignment in bioinformatic field

[huda nawaf]

hi,

I got global seq. alignment code in

http://www.cs.columbia.edu/~cleslie/cs4761/spr04-hw2/spr04-homework2.html

the name of file is align.m.

But need local alignment(smith-waterman algo.)

who can modify the algo. in align.m toe local alignment.

I can design the code to get score, but i faced problem to make algo. display the alignment between two seq.

thanks

[Walter Roberson]

Are you still trying to work with non-biological data? If so then be aware that the Smith-Waterman algorithm is "Identification of Common Molecular Subsequences" by "T. F. SMITH AND M. S. WATERMAN", and is thus inherently about biological data.

It is possible that the techniques used by Smith-Waterman could be adapted for non-biological data, but then it would not be the Smith-Waterman algorithm.

I note that Wikipedia http://en.wikipedia.org/wiki/Smith%E2%80%93Waterman_algorithm claims that the Smith-Waterman algorithm in its original form is not often used now, partly because it is slow.

[Diego]

Did you already look at the swalign.m file located inside the bioinfo/bioinfo folder?.