PSI-MOD Protein Modifications Ontology

This repository contains the Human Proteome Organization (HUPO) Proteomics Standards Initiative (PSI) Protein Modifications Ontology (PSI-MOD). It is the source repository of the PSI-MOD OBO file, main deliverable PSI-MOD.

The PSI working group PSI-MOD

The description of the joint activities of the PSI-MOD working group can be found here.

If you would like to comment on the PSI-MOD document, please submit a new issue through GitHub.

PSI-MOD file formats: OBO, OBO-XML, and OWL

The master source file for this ontology is the OBO file and the OWL file is a derivative of it.

The previously available OBO-XML file format has been deprecated in favor of the more highly supported OWL file format. Pull Requests MUST only update the PSI-MOD.obo file. The ROBOT Tool is used to create compliant OWL files from the OBO using this syntax which is done automatically:

robot.bat convert --input "path-to-obo-file\psi-mod.obo" --output "path-to-obo-file\psi-mod.owl"

The PSI-MOD-newstyle.obo is an automatically generated OBO file that follows the OBO specification closer than the PSI-MOD.obo file. This can best be seen in the use of property_value instead of xref for values on the terms, but there are other changes as well. This file is used for the automatic conversion into the OWL format.

History

The present repository is replacing the previous location at SourceForge CVS site (https://sourceforge.net/projects/psidev/). This link (http://psidev.cvs.sourceforge.net/viewvc/*checkout*/psidev/psi/mod/data/psi-mod.obo) found in various files can be safely substituted with the PSI-MOD PURL: (http://purl.obolibrary.org/obo/mod.obo).

Annotation rules

• Naturally occurring cross-linkers are allowed, but cross-linkers added in the experiment should be added to XL-MOD.
• Modifications should form a tree of increasing specificity which allows users to encode the exact level of specificity known.

oxidized residue -> hydroxylated residue -> hydroxylated arginine -> 4-hydroxy-L-arginine

Mandatory keys

• id the identifier (MOD:XXXXX).
• name
  • Full exact name non ambiguous, usage of common name is allowed (for leaf terms): N4-methyl-L-asparagine and not N-methyl-asparagine
  • Name is the final product (not a reaction description):methylated asparagine and not arginine methylation
  • Can include a precision about origin residue, in parentheses, to remove ambiguity: 2-pyrrolidone-5-carboxylic acid (Glu)
• def describe the modification. After the def any cross identifiers can be listed. These are ids from other databases or ontologies that describe the same modification, or identifiers for papers describing this modification. Common cross references are to PubMed, RESID, ChEBI, CAS, RHEA, uniprot ptm-list. Example patterns for descriptions:
  • A protein modification that effectively converts a SOURCE RESIDUE to MOD.
  • A protein modification that crosslinks two cysteine residues by formation of a chain of two or more bonded sulfur atoms.
  • A protein modification that effectively crosslinks an N-formyl-L-methionine residue and an L-histidine residue to form N-[(L-histidin-1'-yl)methyl]-L-methionine.
  • A protein modification that effectively converts four L-cysteine residues and a four-iron four-sulfur cluster to tetrakis-L-cysteinyl tetrairon tetrasulfide.
• is_a the parent terms. Each entry should be a child of two parent terms, one of each main branch: by amino acid modified (MOD:01157, e.g. MOD:00916 ! modified L-serine residue) and by chemical process (MOD:01156, e.g. MOD:01685 ! alpha-amino palmitoylated residue).

Optional keys

• comment a comment for human readers of the OBO file. If this modification is a cross-linker it needs to start with cross-link NUM to indicate the number of residues involved.
• synonym all synonyms for the term, these are names used in other databases or ontologies, these have to be listed as exact or related. Exact can only be used if it describes a synonym that is exactly the same meaning also exactly as specific. So Unimod synonyms always have to be listed as related.
• relationship used to add relationships other than is_a between terms. Any [Typedef] can be used as a relationship.
• xref see table:

xref	Type	Explanation	Examples
DiffFormula	Molecular formula or none	The difference in molecular formula between the source residue and the modified residue.	C 1 H 2 N 0 O 0 C 0 H -2 N 0 O 0 S 0 (12)C 10 (13)C 4 H 25 (14)N 2 (15)N 1 O 2 S 1
DiffAvg	Number (2 decimal places) or none	The difference in average mass between the source residue and the modified residue.	14.03
DiffMono	Number (6 decimal places) or none	The difference in monoisotopic mass between the source residue and the modified residue.	14.015650
Formula	Molecular formula or none	The full molecular formula of the modified residue (amino acid + modification(s))	C 6 H 10 N 2 O 2 C 3 H 5 N 1 O 1 Se 1
MassAvg	Number (2 decimal places) or none	The full average mass of the modified residue (amino acid + modification(s)).	142.16
MassMono	Number (6 decimal places) or none	The full monoisotopic mass of the modified residue (amino acid + modification(s)).	142.074228
FormalCharge	Number followed by sign	The intrinsic charge contributed by a modification due to its fixed structure (e.g., a permanently charged quaternary amine), excluding any charge arising from pH-dependent (de)protonation or from labile protons or other adducts.	1+ 2-
Origin	string or none	The source amino acid residue(s) from which the modified residue is derived. The DiffFormula, DiffMono, and DiffAvg values are defined relative to the composition of these source residue(s). For canonical amino acids, the one-letter code is used. For non-canonical source residues, the PSI-MOD accession is used instead. For cross-linking modifications, all participating residues are listed, separated by commas.	N C, Q C, C, C, C MOD:00047 H, MOD:00030
Smiles	string	An OpenSMILES string encoding the full modified residue (residue + modification(s)) with * to indicate the backbone bonds.	C(*)(=O)[C@H](COP(O)(=O)O)N*
Source	string	One of natural, artifact, hypothetical, or none, if multiple are applicable take the first from this list.	natural
TermSpec	string	One of N-term, C-term, or none.	N-term

Molecular formula

The elements have to be sorted according to the Hill system. Element numbers can be negative in the DiffFormula. Isotopes have to be encased in round brackets (15) and listed before elements. Elements and element amounts have to be separated by spaces. Some examples: C 0 H -2 N 0 O 0 S 0, (12)C 10 (13)C 4 H 25 (14)N 2 (15)N 1 O 2 S 1.