Sequence based clustering to divide into groups

[atulkakrana]

Hello All,

I have a set of ~1000 sequences, they are of different length, minimum being 500nt. I want separate them into groups on the basis of similarity (phylogeny). I was using clustalw2 to do an multiple alignment and generate phylogenetic tree. But the problem is that manually dividing them into different groups by looking at phylogenetic tree will take a long time and can be error prone.

Is there some tool which I can use to automatically divide (cluster) the sequences into different groups?

I would really appreciate your help. Thanks

AK

[jujubix]

I would recommend using the --cluster option in USEARCH. It's freely available for academic use.

[JackieBadger]

Firstly, all sequences in your phylogeny should be of the same region and thus of the same length.
Second, there are lots of ways to cluster sequences. You need to take into consideration different factors depending on how you sequenced your samples e.g.. error generated via Next Gen.
The metagenomic literature deals with this a lot e.g. http://www.biomedcentral.com/1471-2105/11/152/ or http://www.ncbi.nlm.nih.gov/pubmed/21718538 although I don't believe the metagenomic literature deals with error sufficiently.

Aside from this, there are countless clustering programs out there...google works well