1
$\begingroup$

I have a 120-concatenated protein sequence in a fasta file. I would like to split by proteins. How can I identify each protein?

$\endgroup$
6
  • $\begingroup$ Do you have a multi-fasta file (A FASTA file with multiple sequences, each with its own header) or a mangled FASTA file with one sequence that is in reality 120 individual sequences concatenated? $\endgroup$ – Ram RS Mar 9 '20 at 16:08
  • $\begingroup$ A FASTA file with only one sequence. $\endgroup$ – Manuel Dominguez Becerra Mar 9 '20 at 16:17
  • $\begingroup$ In that case, how do you determine where one protein ends and another begins? How was this file generated? $\endgroup$ – Ram RS Mar 9 '20 at 16:21
  • $\begingroup$ That is what I want to know. The file was generated using GTDBtk, a toolkit programme that identifies 120 genes marker to perform taxonomic classifications. I am using this programme for another purpose. I need the sequence of each protein, but I am afraid I am not going to be able to do it, right? $\endgroup$ – Manuel Dominguez Becerra Mar 9 '20 at 16:33
  • $\begingroup$ sounds really tough unless you know how it was joined in the first place $\endgroup$ – StupidWolf Mar 9 '20 at 16:51
1
$\begingroup$

The only way I can think is to break the protein apart on methionine residues, blast each amino acid locus and concatenate any locus that does not score a query protein position of '1'. The theory is simple every protein starts with a methionine residue, however there are plenty of M residues within the protein that don't represent the beginning of the gene.

Breaking the protein apart on M would be easy, but then automating and parsing the blast would be fiddly. Obvioulsy you will need to code it and if its a one off would it be worth it?

$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.