I have been trying to do clustalx for my sequences of Polyketide synthase. Many of my sequences have two Acyl Carrier protein (ACP) domains. The problem is that clustal x is not accepting the file. However I had tried to change the name of both the domains to avoid ambiguity, I also did BLAST for those domains, no results. I although tried eliminating one ACP domain from the sequences having two which was accepted, but this won\t answer my query. Any help?
Hi Anchal
what do you exactly mean by 'Clustal X is not accepting the file'....are your sequences in the formats recognized by ClustalX....check the formats carefully...there should not be any problem with only MSA of two file...
bye
i did other domain the same way.. it worked... I know there is some probs in my sequence format.. but hv nt been able to figure out wat??/
could you paste the fasta / msa headers of your sequences?
Clustal has a terrible way of handling sequences, that is, if it finds a special character (anything asides from the alphabet or numerals), a blank space or a line break, it will chunk the name of the sequence and it will refuse to accept your files in the case of having duplicates.
For instance, lets assume you have three sequences like these:
>Sequence A
NNNNNNNNNNNNNNNNN
>Sequence B
NNNNNNNNNNNNNNNNN
>Sequence C
NNNNNNNNNNNNNNNNN
for the human eye these three sequences have different headers, but for clustal, it would appear as if you have three times the name "Sequence", it does not handle blank spaces as characters, so the "A", "B" and "C" will be omitted.
Double check your files. I've taught bioinformatics for 2 years and this is the most frequent mistake made when feeding clustal with sequences. Try using underscores instead of blank spaces / special characters. I'm sure it will work.
If it still doesn't work, it might be a problem of your installation, in which case you can always rely on the clustal-x web server, powered by the EBI:
http://www.ebi.ac.uk/Tools/msa/clustalw2/
Hope it was helpful.
Best regards.
Hi Anchal
I also agree with Victor......if you could upload the fasta file...It would be really helpful to figure out the problem you are facing in alignment
bye
The sequence must be FASTA and try with clustal w and check the sequence name not be same
you can convert your sequence to a suitable format appropriate for clustalW, using "Read Seq" program
http://www-bimas.cit.nih.gov/molbio/readseq/
if the problem is ambiquity, clustal stops reading the headers after space so if your sequences are like this
>pks1 ACP 1
NNNNN
>pks1 ACP2
NNNNNNNNN
It wont read after pks1 since there is a space, you can either put an underscore like pks1_ACP_1 or put domains first like ACP1 pks1 and ACP2 pks1 .
I hope this would help
You can also try MEGA, it sees the ambiquity and assigns 1,2 , and so forth after every duplicate sequence.
@Sonal
If I am not wrong many programs has implemented Clustal for MSA...including MEGA...if clustalx is giving some problem..it may give the same problem when performing MSA using MEGA
@Ajay...yes it uses clustalw for alignment, but it doesnt show error for ambiquity of headers and adds 1, 2
like say for example >pks1_acp domain and >pks1_acp domain, clustalw wont accept it but mega first prepares the sequences and during tht time it would add >pks1_acp domain 1 and >pks1 acp domain 2. I had the same problem with clustalw but I used Mega and it helped.
My answer was for ambiquity of sequences, if I understood properly if Anchal is refering to some other problem then I can accordingly
Don't worry use JALVIEW tool firs collect all the sequence in a common .txt file and uploade into jalview your probl;em will solve
@Ajay ....I add an image, the highlighted Yersinia sps has two sequences with same name (check no. 12 and 14). You will get an answer :)
@Sonal
Hi
I guess you are absolutely right...even I did not noticed any error in using 'clustalw' that is related to the sequence headers....and thanks for loading the sequence file for clarifying the same....as you pointed out...Anchal must be facing the error due to some other problem..may be in the formatting or some invalid character...only if she can upload the seqeunces people can give some suggestions...
bye
Still It would be helpful if she uploads her data so that we stop blind-guessing what might be the problem and provide a simple and straightforward solution to her problem.
Thanks for the suggestion ppl. I wud try to upload the sequences...although uploading 300 sequences is a tedious job.... I am also luking for another alternative to this prob as well
Just the fasta headers. we are pretty sure is not an issue of the sequences but of the fasta/msa headers
- Badges
- Science topic
- Similar topics
- Bioinformatics
- Bioinformatics and Computational Biology