ClustalO computing suboptimal alignment

Question

[mschmid]

Clustal Omega - 1.2.4

I align about 600 (n) highly identical sequences. n-1 sequences are identical. There is one sequence which has some differences. For some regions I get an alignment which is obviously suboptiomal (gap is one base to early).

EDIT: This is just one example. Obviously I could just correct this manually. But I automate a process where such alignments should not happen.

ClustalO call is the following:

clustalo -i inp.fasta -o out.clo -t DNA --outfmt=clu --full --dealign --auto

I get the following suboptimal alignment (just a small part not showing all the n-1 identical sequences):

CTCAAGTAACTTCTTTT
CTCAAGTAACTTCTTTT
CTCAAGTAACTTCTTTT
CTCAAGTAACTTCTTTT
CTCAAGTA-ACTTCTTT
********   *  ***

How can I:

1. Optimize the aligment with ClustalO?
2. Do a post-optimization?
3. Or do I have to use a different aligner? For mostly highly identical sequences.

[Joe]

Any aligner should have spotted that, but its not uncommon to see things like this.

I would try with a different aligner (Clustal typically tends to be better with protein sequences).

Why are you using --dealign?

[mschmid]

Yes, MUSCLE seems to do a much better job for my application. Strange though....

Answer 1

[mschmid]

I tried MUSCLE and it seems to do a much better job.