This slot applies to both small molecular weight compounds, and to proteins. In the case of proteins, it is used to encode chemical modifications to proteins, such as methylation. In this case, one of the ATOMs would be the ID of a protein frame.

reference-id = a PubMed/Medline identifier (we prefer the PubMed id) or id for a Publication frame (w/o the leading "PUB-") evidence-code= the frame id of some subclass of |Evidence|, e.g. EV-EXP timestamp = a lisp universal time corresponding to the time the evidence code was assigned curator = the username of the curator who assigned the evidence code probability = a number between 0 and 1 describing the probability that the evidence is correct

Any of the above components may be omitted, but it is meaningless to supply a timestamp, curator or probability if the evidence-code is omitted. Trailing colons should be omitted, but if a value contains an evidence-code with no accompanying citation, the leading colon must be present. The square brackets are optional.

Examples: [123456] -- a PubMed or Medline reference [SMITH95] -- a non-PubMed reference [123456:EV-IDA] -- an evidence code with associated PubMed reference [:EV-HINF] -- an evidence code with no associated reference [123456:EV-IGI:9876543:paley] -- a time- and user-stamped evidence code with associated reference [123456:EV-AINF:9876543:paley:0.75] -- a fully instantiated evidence code and reference

When a modified protein is created to reflect a chemically modified alternate form of the protein, if that protein is a complex, its subunit structure need not be mirrored in the modified-protein frame. The subunit structure should be defined only once, for the frame that represents the unmodified form of the protein.

Each value of this slot is a tuple of the form (R ER1 ... ERn) where R is a reaction frame ID, and the one or more ERi are the IDs of enzymatic reaction frames. A given tuple specifies that the enzyme linked to ER1 catalyzes R.

When a polypeptide exists in one or more chemically modified forms, all of those forms of the polypeptide should link to the same gene frame using the Gene slot.

1. Standard 2. Vertebrate Mitochondrial 3. Yeast Mitochondrial 4. Mold, Protozoan, Coelenterate Mitochondrial and Mycoplasma/Spiroplasma 5. Invertebrate Mitochondrial 6. Ciliate, Dasycladacean and Hexamita (no tables 7 or 8) 9. Echinoderm Mitochondrial 10. Euplotid Nuclear 11. Bacterial (same as standard, except for different alternate initiation codons) 12. Alternative Yeast Nuclear 13. Ascidian Mitochondrial 14. Flatworm Mitochondrial 15. Blepharisma Nuclear

See http://anatomy.med.unsw.edu.au/CBL/embryo/DNA/Genetic_Codes.htm for more info, including systematic ranges for each code and citations.

:CYCLE-TOP-CPD -- The advice is a compound key. In pathways containing a cycle, the cycle will be rotated so that the specified comound is positioned at twelve o'clock.

:CASCADE-RXN-ORDERING -- The advice is a list of reactions that form a partial order for reactions in a cascade pathway (i.e. the 2-component signalling pathways).

Please cite XXXCyc as CITATION in publications resulting from its use.

Possible values of this slot are:

REVERSIBLE: Reaction occurs in both directions in physiological settings.

PHYSIOL-LEFT-TO-RIGHT PHYSIOL-RIGHT-TO-LEFT The reaction occurs in the specified direction in physiological settings, because of several possible factors including the energetics of the reaction, local concentrations of reactants and products, and the regulation of the enzyme or its expression.

IRREVERSIBLE-LEFT-TO-RIGHT IRREVERSIBLE-RIGHT-TO-LEFT For all practical purposes, the reaction occurs only in the specified direction in physiological settings, because of chemical properties of the reaction.

This slot is used by the PathoLogic code that does schema copying, and by the code in kb-update.lisp that does schema mirroring.

The general rule is that classes that should be copied from MetaCyc to a new organism by PathoLogic are considered schema classes. Examples include the hierarchies of compounds and pathways, and general classes like Genes and Proteins and All-ACPs. Examples of classes to exclude are classes for specific modified proteins that occur as substrates in reactions, such as Thioredoxin. The problem with copying Theoredoxin is that the reactions it appears as a substrate in are not copied when the class is copied, causing bad references to those reactions.

Put another way, no protein should have both an unmodified form and a modified form; every modified protein should be linked back to the base form of the protein.