CH$LINK: clarification

[meowcat]

Hi,

in the record format, it is not clearly specified how to handle potential multiple entries for one CH$LINK: DATABASE, e.g. a link to multiple CAS numbers. Can they be added as multiple entries and/or as space-separated entries? I think both would pass the validator. I think the former would be better.

https://github.com/MassBank/MassBank-web/blob/main/Documentation/MassBankRecordFormat.md#228-chlink-subtag-identifier

[schymane]

I would be pro multiple entries ... but ... there's only one CAS number per structure technically ... and the others are "alternative forms". So strictly correct would be to only report the CAS matching by InChIKey ... but .. the standards used are often the salt form which is a different CAS. Do we have to "comment" these differences somehow? (or should we introduce the ability to do so?)

Just some examples (I've also seen "related CAS" used)

https://pubchem.ncbi.nlm.nih.gov/compound/2256#section=Other-Identifiers

https://comptox.epa.gov/dashboard/dsstoxdb/results?search=DTXSID9020112#synonyms

[tsufz]

@schymane and @meowcat, I suggest that you develop a proposal for the MassBank meeting. I would like to go ahead with the preferred MassBank name anyway and other standardization issues can be part of this implementation.

[meowcat]

The purpose I want to use this for is not actually CAS but a linking system I am working on for (currently) internal use (see #303); CAS was just an example to make clear that this case is not specified.

[meowcat]

(I've also seen "related CAS" used)

I think that's what NIST does IIRC?

[schymane]

I thought ChemSpider and some of the other DBs too but could not find quickly ... maybe I picked a bad example in atrazine ...