which structure should be used for a drug (like fluvastatin) in chemoinformatics

Question

[Zhilong Jia]

For drug Fluvastatin, Fluvastatin [sodium salt] (CAS-93957-55-2) is usually used for testing the potential of Fluvastatin. They have different structures, though it's slight difference.

In Drugbank, the Fluvastatin structure is at the beginning, while the salt form in the latter section of the webpage.

It seems the active form should be Fluvastatin ion in cell.

So which structure should be used when considering the structure ot this drug in chemoinformatics? Thank you.

Update:

Another example:

For tamoxifen, one of its Synonyms is tamoxifen citrate shown in guidetopharmacology. In pubchem, they are different (CID 2733526 and 2733525) with the latter includes a citrate. At least in the Connectivity Map, tamoxifen citrate [54965-24-1] (as a catalog name) was used instead of tamoxifen (as a cmap_name) [ref1]. I think they are just different, resulting in different effect on cells. In this situation, should I use the structure of tamoxifen to discover something in the Connectivity Map? It's a little odd. I'm wondering why they use tamoxifen citrate instead of tamoxifen, but name it as tamoxifen. Is it common for drug testing in chemistry?

[cdsouthan]

What do you actually want to do? Connectivity Map substance mapping is useful (you can order the same stuff they used for their experiment) but also problematic (its not always possible to make a clean mapping between catalogue number/name/CASno and a standardised molecular stucture entry in PubChem or other databases)

[Zhilong Jia]

Yes, The mapping for Cmap data is difficult. I aim to do sth. related the structure of the compounds used in cmap. Just wondering why cmap or chemist test some compounds (like tamoxifen citrate) but not drugs (like tamoxifen) themselves for the potential of drugs. Is it common in chemistry? Or it's just a mistaken in cmap. I believe there are some difference in these two compounds based on your reply.

[cdsouthan]

This gets complex but its the difference between cheminformatics and the essentially biochemical experiments in Cmap. For example, since they dissolve in DMSO and then dilute extensively in buffer it wont really matter which counterion was in the drug solid form they bought (except needing to know it to weigh out for the concentration calculation) so the bioactivity (e.g. expression changes) can be primarily attributed to the parent structure (and its metabolites). However, the USAN and FDA need to indicate that, for example, tamoxifen citrate or flouvastatin calcium is in the tablets for patients.

[Zhilong Jia]

I found tamoxifen citrate is the main composition of drug tamoxifen. So, using the structure of tamoxifen is fine in cheminformatics concerning the CMap situation. Thank you.

[cdsouthan]

Thanks for the spot so I have removed tamoxifen citrate as a (misleading) synoym in http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=1016 (but the change wont go live until our new release in a couple of weeks)

Answer 1

[cdsouthan]

You have chosen a tricky one because it has salts, racemates, enantiomers and virtual deuteration

this adds up to 113 different representations http://www.ncbi.nlm.nih.gov/pccompound?cmd=Link&LinkName=pccompound_pccompound_parent_connectivity_pulldown&from_uid=1548972

Choices depends on exactly what cheminformatic operations you intend to perform, but stripping salts back to parents is usually advisable

To keep life simple just go with this GtoPdb entry

http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=2951

[Zhilong Jia]

Sometimes, drugs used in cell experiment include some other groups, not as simple as a salt. See Update please. Thank you.