Ignore Error "Multiple Sequences Found With Same Name" In Clustalw

Question

[david]

Hi,

I have a python program generating a clustalw2 alignment of about 500 sequences from a fasta file. The names of the sequences correspond to the respective organisms plus the substrate specificity of a given sequence. Therefore quite a few of these names are identical and i get the error message: "Error: Multiple sequences found with same name" and no alignment is generated. Is it possible to ignore this error without having to change all the sequence names?

Cheers David

Answer 1

[Andrzej Zielezinski]

The names of the sequences must be unique to do alignment in ClustalW/X.

I would name your 500 sequences as numbers from 0 to 499 and store the original names in a dictionary or a list.

For example:

d = {1: 'Organism1Substrate', 2:'Organism1Substrate' , ..., 499:'Organism2Substrate'}

or:

l = ['Organism1Substrate', 'Organism1Substrate', 'Organism1Substrate', ..]

Once you performed the alignment, just replace the numbers with original names.

[Josh Herr]

+1 for this. In the past I have just GREPed the names and added numbers or more information to make them unique, but I like this idea better.

[qiyunzhu]

Agree. Many phylogenetic programs have problems handling fancy sequence names. The horrible case is phylip format (used by RAxML etc) which allows only 10 characters per name. So I always rename the sequences as "s1", "s2", s3"... I don't recommend using 1, 2, 3... because some programs cannot handle numerical sequence names.