Moe, Sorry about the late reply. I was traveling in the past few weeks.
The documentation is probably not clear, but what you observed is the intended behavior. The -cD option indeed re-maps the peptides to the database you specified, but it only changes the "Protein=" field. The NISTProtein= field is copied from the original NIST library (just to keep a record), and is not touched by SpectraST. The -cu and -cd options do NOT just remove the unmapped/multimapped protein names from the "Protein=" field. They remove the library entry entirely. (Again, it does not look at the NISTProtein= field at all.) The purpose of the options is to update libraries when protein sequence databases change, or to trim libraries to "sub-libraries" containing only the proteins you want, not to edit the Comment in the library entries. If you want to edit the Comment field (e.g. to remove the NISTProtein field), you can safely do so with a text editor (or a script if you want to do it automatically) on the .sptxt file, rename it to .msp, and re-import. I don't see why this is useful though unless you are using the library sptxt file as a storage or an intermediate file which will be read by your own program. Henry On Wednesday, July 11, 2018 at 6:25:54 AM UTC+8, ML wrote: > > Hi all, > > I'm new to TPP and I'm intending to update the Rosenberger panhuman > library from SWATH atlas to the latest uniprot/swissprot release. > *I want to unmap non-existing proteins and only retain newly mapped > proteins*. > > If you want to rework the issue: > The phl004_consensus.sptxt file can be downloaded here > https://db.systemsbiology.net/sbeams/cgi/PeptideAtlas/GetDIALibs but I've > also attached a small example file (single peptide copied and pasted from > the panhuman library).The uniprot fasta file can be downloaded also here > https://www.uniprot.org/uniprot/?query=organism%3A%22Homo+sapiens+%28Human%29+%5B9606%5D%22+NOT+existence%3Auncertain+reviewed%3Ayes > > In the small example: > During remapping with spectraST the old uniprot identifier remains in the > library file (e.g. P06311) and the newly remapped from the fasta file get > appended. > *I expected that P06311 would get deleted during the remapping as it does > not exist in the fasta file.* > > For example, the peptide in question (ASQSVSSNLAWYQQK in the panhuman > library is mapping to P06311) however, this protein doesn't exist anymore > and the sequence of ASQSVSSNLAWYQQK should now map to 2 proteins, namely > A0A087WSY6 and P01624. > > I'm using the following command to import the library and to create new > files including: imported.splib and imported.sptxt > *spectrast -cNimported small.sptxt * > > #When I look at imported.sptxt in this "original" state it is the same as > in the "small.sptxt" file, the ASQSVSSNLAWYQQK peptide maps to > NISTProtein=1/P06311, FullName: X.ASQSVSSNLAWYQQK.X/2 (CID) > > followed by > *spectrast -cD20180704_9606_rev_uniprot.fasta imported.splib* #the > peptide now maps to NISTProtein=1/1/P06311 and to > Protein=2/sp|A0A087WSY6|KVD15_HUMAN/sp|P01624|KV315_HUMAN, FullName: > R.ASQSVSSNLAWYQQK.P/2 (CID) > > I thought I could use the -cu flag (refreshDeleteUnmapped) to remove the > original mapping but it doesn't seem to do that, the output seems identical > to not using the -cu flag: > * spectrast -cD20180704_9606_rev_uniprot.fasta -cu imported.splib* > > > When I use the -cd flag (refreshDeleteMultimapped), the whole peptide gets > deleted - presumably due to multiple mappings > *spectrast -cD20180704_9606_rev_uniprot.fasta -cu -cd imported.splib* > > Any help would be appreciated. > > Regards > Moe > > > -- You received this message because you are subscribed to the Google Groups "spctools-discuss" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/spctools-discuss. For more options, visit https://groups.google.com/d/optout.
