anesvi
commented
4 years ago In PSM.tsv.ion.tsv/peptide.tsv, all mapped proteins for that sequence are shown in Mapped column (Protein column shows where it is razor)
In Protein.tsv, there is Indistinguishable protein column, showing all other proteins when several proteins cannot be distinguished from the one shown in Protein column
If you have a shorter isoform, but the longer one got an extra peptide, then the shorter one will not be listed in Indistinguishable. You will see it in the peptide-level files, but not in the protein file. But if the shorter and the longer isoform are identified by exactly the same peptides, you should see them both. One will be in Protein column, and the other in Indistinguishable column
Alexey
From: JB91451 notifications@github.com Sent: Tuesday, December 10, 2019 7:35 AM To: Nesvilab/philosopher philosopher@noreply.github.com Cc: Subscribed subscribed@noreply.github.com Subject: [Nesvilab/philosopher] Protein grouping in report tables (#93)
External Email - Use Caution
Dear Felipe,
I'm not sure about the protein grouping that is applied in the different reports from philosopher (or maybe this question is more related to ProteinProphet). In one of our current samples I found the following: In the peptides table there are 6 peptides that belong to protein A in the main column and are also mapped to an other protein B (actually A is just a truncated version of B and there should be no way to distinguish them in normal proteomics). However, in the protein table there is only protein A given and the mapped protein column is empty.
Is this the intended behaviour as long as there is no evidence that protein B could also be present? Is it only the fact that protein A is shorter and thus can explain all peptides with a higher sequence coverage than B would and will this change when I use the "noOccam" option? Unfortunately our database contains a lot of such cases as it contains a 6-frame translation with different start codons.
Best regards, Juergen
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/Nesvilab/philosopher/issues/93?email_source=notifications&email_token=AIIMM64DQFH5FBP2G6RB35TQX6EI7A5CNFSM4JY5YVL2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4H7OFEJQ, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AIIMM65I55QJLDUSDAYEOB3QX6EI7ANCNFSM4JY5YVLQ.
Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues
JB91451
Dear Felipe,
I'm not sure about the protein grouping that is applied in the different reports from philosopher (or maybe this question is more related to ProteinProphet). In one of our current samples I found the following: In the peptides table there are 6 peptides that belong to protein A in the main column and are also mapped to an other protein B (actually A is just a truncated version of B and there should be no way to distinguish them in normal proteomics). However, in the protein table there is only protein A given and the mapped protein column is empty.
Is this the intended behaviour as long as there is no evidence that protein B could also be present? Is it only the fact that protein A is shorter and thus can explain all peptides with a higher sequence coverage than B would and will this change when I use the "noOccam" option? Unfortunately our database contains a lot of such cases as it contains a 6-frame translation with different start codons.
Best regards, Juergen