help on a simple (?) assembly task

[Kennels]

Hi,

I've never done an actual assembly before so I'm hoping someone can give me some advice to a presumably simple task.

I've got some 100base read paired end genome sequence data of an organism where the genome sequence is not known. At the moment, I have a gene of interest in a closely related organism, and all I want to do is use this gene of interest sequence as a template to:

1.Pull out closely matching reads from my data using Bowtie
2.Use those reads to create a scaffold using velvet (?)
3. Use this initial scaffold to then just extend a couple of kb on either side, again with velvet (?)

Basically, I just want to deduce the orthologue with 5',3' UTR from the dataset.

Is my workflow correct?
Should I be using other software?
Can velvet be set to limit assembly to within a defined range from an initial contig/scaffold?

Thanks for any advice.

[Kennels]

anyone have any sort of advice?
thanks.

[gaffa]

An alternative approach could be to assemble all of your reads first, and then on the contig-level search for (e.g. by BLAST) sequences similar to your gene of interest.

[Kennels]

thanks for the reply.

That was my first thought but unfortunately I am limited in computing power (6cpus and only 4 gigs of RAM) at the moment and thus need kind of a clumsy way of doing it.

I've actually managed to use bowtie to identify reads aligning to this reference, and then input these into velvet to produce a couple of contigs.

What I'd like to do is now use these contigs as an anchor to just extend out the assembly for a couple of kb's, but I can't figure out an option to do this in velvet, or if I should take a different approach now.

Any more suggestions?
cheers