Multiple Sequence Alignment using DNA sequence
0
1
Entering edit mode
bbista91 ▴ 10
@bbista91-13099
Last seen 7.6 years ago

I want to align a DNA sequence by first translating it to a protein rather than aligning the DNA itself. Then using msa() I would align the sequences but how would I convert it back to the original DNA sequences aligned on the basis of the AAs.

msa decipher • 2.1k views
ADD COMMENT
1
Entering edit mode

Hi,

This is an interesting problem. I'm curious to hear from Erik Wright, the author of the DECIPHER package, about this. IIRC Erik is doing something like this in DECIPHER to efficiently align/compare DNA sequences. The "reverse-translation" step is clearly the challenge here, given that the mapping from codons to amino acids is many-to-one. My guess is that one would need to keep track of which codon was translated into which amino acid, for all translated codons. Then "reverse-translation" should be straightforward. However it seems to me that, when doing this, one cannot and should not expect that 2 well aligned AA sequences will "reverse-translate" into 2 well aligned DNA sequences in general.

Can you please tag your question with DECIPHER?

Thanks,

H.

ADD REPLY
0
Entering edit mode

I am new to Bioinformatics but I looked at the DECIPHER package and it does do what I want with ease. My only concern now is how reliable is the Alignment using DECIPHER, my supervisor prefers MUSCLE. Do you have any thoughts on that?

Thank you so much for pointing me in the right direction.

B

ADD REPLY
0
Entering edit mode

Hello,

Sorry for the delay - I did not see the original post.

Indeed DECIPHER will perform alignment in amino acid space on DNA/RNA coding sequences.  Conceptually it is equivalent to translating, aligning, and then reverse translating as Hervé says.  But it does not actually need to reverse translate, as all DECIPHER needs to know is where the gaps were added in the amino acid sequences and it can apply equivalent gaps to those codon positions.

As for the question of what is the accuracy in comparison to MUSCLE...  You should take a look at this paper. As you can see, DECIPHER outperforms MUSCLE and other alignment programs, especially as the number of sequences being aligned increases.

I hope that answers your question, and sorry again for the delayed response.

Erik

ADD REPLY

Login before adding your answer.

Traffic: 592 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6