some "not great" ntrs from the iedb

[rvita]

Hi, we have a few term requests that are not exactly protein modifications, but we use them in our data in a similar way that we use PSI-MOD terms, so we are hoping that you can help us with them. If the are not appropriate for PSI-MOD, we hope you can suggest a different home for them. They are: D-aminoacid Stereoisomer I/L uncertainty (when eluted peptide are sequenced, this is a common issue) Unidentified amino acid (when eluted peptide are sequenced, this is a common issue) artificial modification/molecular tag - is this already a concept in PSI-MOD? Iodo-4-azidosalicylic acid (IASA) is an example that we have. Please let me know which, if any are within your scope and I can work towards definitions, etc. thanks, Randi

[pabinz]

Hi Randi, PSI-MOD is not restricted to natural PTMs. Therefore there is space for a number of other mods. And we have a number of them already. D-aminoacid : Yes, can be there Stereoisomer: Yes, can be there I/L uncertainty (when eluted peptide are sequenced, this is a common issue); a kind of ambiguity, yes in principle, a special case. Unidentified amino acid (when eluted peptide are sequenced, this is a common issue): full ambiguity with no mass value. We should have had this since the beginning of the creation of ambiguous terms, to be added artificial modification/molecular tag - is this already a concept in PSI-MOD? Iodo-4-azidosalicylic acid (IASA): yes and we have already many of them (itraq for instance)

I let you propose definitions for your new terms

Thanks Pierre-Alain

De : rvita @.> Envoyé : mercredi 30 mars 2022 18:12 À : HUPO-PSI/psi-mod-CV @.> Cc : Subscribed @.***> Objet : [HUPO-PSI/psi-mod-CV] some "not great" ntrs from the iedb (Issue #67)

Hi, we have a few term requests that are not exactly protein modifications, but we use them in our data in a similar way that we use PSI-MOD terms, so we are hoping that you can help us with them. If the are not appropriate for PSI-MOD, we hope you can suggest a different home for them. They are: D-aminoacid Stereoisomer I/L uncertainty (when eluted peptide are sequenced, this is a common issue) Unidentified amino acid (when eluted peptide are sequenced, this is a common issue) artificial modification/molecular tag - is this already a concept in PSI-MOD? Iodo-4-azidosalicylic acid (IASA) is an example that we have. Please let me know which, if any are within your scope and I can work towards definitions, etc. thanks, Randi

— Reply to this email directly, view it on GitHubhttps://github.com/HUPO-PSI/psi-mod-CV/issues/67, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AESGSRSPZP7QBUZB3SYYQ6TVCR4MPANCNFSM5SCOJWSA. You are receiving this because you are subscribed to this thread.Message ID: @.**@.>>

[rvita]

ok, for the new terms, here is an attempt to define them. D-amino acid Definition: Amino acid where the stereogenic carbon alpha to the amino group has the D-configuration. Taken from https://en.wikipedia.org/wiki/D-Amino_acid#:~:text=D%2DAmino%20acids%20are%20amino,nature%20as%20residues%20in%20proteins.

Isoleucine/Leucine uncertainty Definition: An experimental outcome whereby isoleucine and leucine residues are indistinguishable because both are characterized by a 113 Da mass difference from the other peptide fragments in the MS2 spectrum. Taken from: https://doi.org/10.1016/j.jasms.2006.08.011

Unidentified amino acid Definition: When the identity of an amino acid is fully ambiguous with no mass value.

Artificial modification/molecular tag Here, I am not sure which term is best to request nor how to define it. The concept of "artificial" could be ascribed to many naturally occurring ptms when artificially induced in a lab setting. The concept of "molecular tag" is dependent upon the reason for incorporating the modification and can apply to many ptms in different settings. We do still want the term, but I am afraid to make a request that is too vague, so any guidance on what works best for you is much appreciated.

Thank you

[nataled]

With respect to 'artificial': This is something that was considered when we started the Protein Ontology (which heavily uses PSI-MOD). Our view is that anything that is made artificially that exactly mimics something natural would be considered natural. That view reduces the complexity of the issue somewhat, though not fully (there's still the possibility of creating what is thought to be an artificial PTM that is later found to be natural).

@pabinz when you say that this is already a concept in PSI-MOD, do you refer to the 'source' when that is given as 'artifact'? If so, would all these then become children of the new term? I think that would be quite useful, and could be achieved automatically with reasoning (in which case the term could be logically defined as 'protein modification and source=artifact'; though I'm not sure if the xref line can be used for reasoning--perhaps with oboInOwl:hasDbXref?).