Continuing with the theme of "insufficiently geeky" today...

[baratron]

Today I am trying to work out whether (deep breath):
(4E,6Z,8S,9S,10E,12S,13R,14S,16R)-13-hydroxy-8,14,19-trimethoxy-4,10,12,16-tetramethyl-3,20,22-trioxo-2-azabicyclo[16.3.1]docosa-1(21),4,6,10,18-pentaen-9-yl carbamate
is the same thing as:
(4Z,6Z,8R,9R,10Z,12R,13R,14R,16S)-13-hydroxy-8,14,19-trimethoxy-4,10,12,16-tetramethyl-3,20,22-trioxo-2-azabicyclo[16.3.1]docosa-1(21),4,6,10,18-pentaen-9-yl carbamate.

It would be easy if all the stereochemistry was inverted, but it isn't - one double bond is the same and one side-chain is the same in both naming schemes.

Anyone who says that I'm a chemist so I should be able to work this out will be shot, since most people's brains freeze up when confronted with a molecule name on three lines. It takes some sort of special IUPAC knowledge to be able to cope with ridiculously complex molecules. Everyone else calls this thing geldanamycin.

The real question is whether I can invert the stereochemistry at every single one of those stereocentres, or if the groups always have to stay in the same configuration relative to each other. I have no clue.

I predict that my evening is going to involve making this monster out of my molecular modelling kit, since I simply can't deal with complicated shape stuff on paper.

Comments

[brooksmoses]

Wow. My immediate take on that is that the only way "you should be able to work it out" could reasonably apply is with the caveat "with the right tools". The molecular modeling kit seems like the right approach, indeed.

I can easily imagine that there are people who believe they could work this out in their head, but I have a much harder time imagining that a significant fraction of them would get the correct answer all the time on such things.

[stellarwind.livejournal.com]

I think my brain just had a seizure trying to imagine this molecule.