Database Update (Standard InChIKeys)
We ran an update of the Chemical Identifier Resolver database tonight which grew the number of chemical structures known by the service quite a bit. The number of Standard InChIKeys indexed with their respective full structure representations is now approx. 93 million (92,939,226 to be exact). The service uses this index to work as an Standard InChIKey Resolver, e.g.
http://cactus.nci.nih.gov/chemical/structure/InChIKey=RZVAJINKPMORJF-UHFFFAOYSA-N/file?format=sdf http://cactus.nci.nih.gov/chemical/structure/InChIKey=RZVAJINKPMORJF-UHFFFAOYSA-N/file?format=cdxml http://cactus.nci.nih.gov/chemical/structure/InChIKey=RZVAJINKPMORJF-UHFFFAOYSA-N/smiles
I am trying to get this to work. I don’t understand if I am using a faulty inchikey or not.
FOSBUCGGRDVOKR-UHFFFAOYSA-N
this one works,
WNYOFUWGTYHVTB-UHFFFAOYSA-N
but not this one:
OAYHOICQTNCZLJ-UHFFFAOYSA-N
the only differecne is one extra carbon
am I doing something wrong?
is there any way to validate an inchikey?
Hi Elliot,
InChIKeys don’t carry any structure information, they are just a hashcode representation of a chemical structure, i.e., whoever draws a specific chemical structure in whatever way the InChI library calculates an unique (Standard) InChIKey from it. However, from the InChIKey itself there is no way back to the structure.
So, to resolve a InChIKey back into a structure the Resolver has to do a lookup in its database which stores the InChIKeys and their corresponding structures. Currently, the database of the Resolver knows about 93 million Standard InChIKeys. So the “problem” is that your first InChIKey is know in the database the second one unfortunately is not