Cluster similar sequences

[shadow19c]

Hello,
I have some sequences which are mapping the end of of referential sequence, and i'm waiting to have 4 groups of different sequences.
So I want to know how can I clusterize similar sequences, eventually using a de novo method of clustering , any suggestions....


Thanks

[gsgs]

many sequences ? fast ?
compute the distance matrix
cumulative nearest unused neighbor
traveling salesman
cumulative nearest distance between grouped and still ungrouped sequences

[suknamgoong]

CD-Hits

http://weizhong-lab.ucsd.edu/cd-hit/

USEARCH/UCLUST

http://www.drive5.com/usearch/

[shadow19c]

Hello,
I want to catch new TEs insertion.
So I sarted with unammped reads, and I tried to force the mapping only in the TEs genome.
I did it with BWA
bwa index -a bwtsw TE_a.fasta
bwa bwasw TE_a.fasta unmapped_reads.fasta > TE_a.sam

I took after only the reads which mapped at the extremity l
cat /archives/a2e/kassam/Epiril260/epiril260_vandal.sam | awk 'and($2, 4)!=4 && $6~/[0-9][0-9]+S/'

And i tried with Cd-hit and Usearch to clusterized the reads.

But I did not have good clustering reads, so I want to know if someone has another idea to do that or methodology?

Thanks a lot