search

MassBank / MassBank-data

Official repository of open data MassBank records
74 stars 59 forks source link

Submit record to Massbank EU #54

Closed zzjl20 closed 5 years ago

zzjl20 commented 5 years ago

The record contains MS and MS2 records. Already checked by .scripts/validate.sh Contact me by: donghan-l@nig.ac.jp

meier-rene commented 5 years ago

Thank you for this contribution. In principle this looks quite good, but I have one issue. Although these records pass the validator they have some problems. In the "COMMENT" section i can find information like "COMMENT: Origin: Animal, CSID: 10334, SubCategory_DNP : Lipids, CASID (tmp): [18951-77-4], Fatty acids". Better would be to place things like "CSID: 10334" and "CASID (tmp): [18951-77-4]" in the CH$LINK section. Please check CH$LINK I would rather see it in this way:

CH$LINK: PUBCHEM CID:10334
CH$LINK: CAS 18951-77-4

in the recordfile.

Do you think it would be possible to create the records in this way? Do you use your own software to create the record files?

meier-rene commented 5 years ago

Nice work. Thank you for your contribution.

meowcat commented 5 years ago

There's still problems here. Things listed as CH$LINK: PUBCHEM CID should really be CH$LINK: CHEMSPIDER instead! At least for the ones I checked.

meier-rene commented 5 years ago

Yes @meowcat, you are right. I tested some and it was never correct. Automatic validation doesn't check this yet. Implementing test for this is on the roadmap, but I'm already a bit afraid of the numbers of mistakes I need to fix...

meowcat commented 5 years ago

I think it will be hard to validate this strictly, since there are multiple true and half-true answers sometimes (stereoisomers, mixtures, salts etc will all not have a simple answer).

schymane commented 5 years ago

It should be easy to check, the ChemSpider ID and PubChem CID should be an InChIKey match, at the very least an InChIKey first block match. Everything else is clearly wrong. Entries that fail an InChIKey check should be validated. Could this case be a misassigned identifier?

zzjl20 commented 5 years ago

OK. thanks for your advice so much. I will check the list.I will try my best to upload high quality records.

https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail ウイルス フリー｡ www.avast.com https://www.avast.com/sig-email?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

Emma Schymanski notifications@github.com 于2019年4月16日周二 上午2:10写道：

It should be easy to check, the ChemSpider ID and PubChem CID should be an InChIKey match, at the very least an InChIKey first block match. Everything else is clearly wrong. Entries that fail an InChIKey check should be validated. Could this case be a misassigned identifier?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/MassBank/MassBank-data/pull/54#issuecomment-483338529, or mute the thread https://github.com/notifications/unsubscribe-auth/AZvoyYodT6XRvz7-aOrXSueydpnk-mI6ks5vhLKIgaJpZM4cU2Qm .