Hey everyone. I'm currently trying to replicate what was done in this paper (Article MSFragger: ultrafast and comprehensive peptide identificatio...
) to get a better handle on how to use MSFragger as a tool for analysis in mass-spec. I'm pretty much a newbie when it comes to mass-spec. I am having trouble designing a fasta database for the program to use.In the methods, it says that " The searched sequence database was created from the human protein sequences of Ensembl version 78 appended with reversed protein sequences as decoys and common contaminants ."
The part that confuses me is the "appended with reversed protein sequences." Should I be taking every single uniprot protein sequence and appending the reverse peptide sequence in this fasta file? I appreciate any help
Cheers,
Yousuf
Hi Yousuf
It depends on what software, you are using. I can only comment on MaxQuant, but here, a reverse decoy database is automatically created during data processing. In principle, it takes your proteins, digests them with Trypsin (or whatever enzyme you are using) and then reverses and then reverses the sequence except the termini. It does so to control the FDR.
Including common contaminants is an option that is selected by default. By identifying these, you are able to go deeper in your data and boost identification.
I have used Proteome Discoverer and MaxQuant. In both case, once you upload your protein database containing contaminants, the decoy database is created on its own. In case of Proteome Discoverer, once you upload the database, there is a option of index the database. You have to go and choose your proteolytic enzyme (eg. trypsin) and the algorithm takes care of the decoy database. Dont forget to restart the Proteome Discoverer after you upload/index your database or else it wont be reflecting your search algorithm (SequestHT or Mascot)
Thanks to both of you. Your answers, along with some more online reading, gave me a much better idea of decoy databases. For anyone who sees this question in the future and wants a program to make a custom decoy database (which is only necessary for programs like MSFragger, not Maxquant or proteome discover which both make it for you automatically), use this
https://www.sanger.ac.uk/science/tools/decoypyrat
Hi Yousuf Khan, I found this step by step which can also help:
https://github.com/Nesvilab/philosopher/wiki/How-to-Prepare-a-Protein-Database
It all begins, starting the "philosopher" program, e.g.:
C:\WINDOWS\system32>philosopher.exe
Your "decoy or reverse sequences" will be in the same folder as "philosopher" (.fas archive).
Regards!
César
- Badges
- Science topic
- Similar topics
- Filing