This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Clustering based on sequence similarity

I've got 5 from 7 to 30 thousand virus genome sequences per each strain and I need to separate the sequences into groups based on the similarity of the sequences. How can I do that? By the way I'm able to align each strain with MAFFT, but i don't really know the way to cluster. I'd be really happy ot hear the answer

relativenessbasedclustering
• 832 views • link
updated by Mensur Dlakic Mensur Dlakic • written by Мики-рики-тави • Мики-рики-тави

I realize it is your username, but I feel like it should be riki-Miki-tavi.

• 0 views • link
written by Mensur Dlakic Mensur Dlakic

1 answer

This part is not clear:

I've got 5 from 7 to 30 thousand virus genome sequences

You have 5 protein sequences from 30K genomes? 5-7 protein sequences? 5-7 genes?

If you are talking about whole genome clustering, that would not be easy on such a scale. I recommend that you use predicted proteins for each of them. Then:

  • align them individually
  • trim the alignments
  • concatenate those alignments into a super-matrix
  • make a phylogenetic tree

Beware that each of these steps, especially the last one, will take a long time. Also, there is a large potential for error when working on this scale, even for those who have already done all these steps before. Even if all of this works, it is very difficult to look through a tree that has 30K nodes. Lastly, most of your genomes will be (near-)identical at a protein level, so you still may not get much useful information.

• 0 views • link
written by Mensur Dlakic Mensur Dlakic

Log in to answer this question.

More posts like this