« Funding in the EU: The Simple Way |
Main
| Another Alzheimer's Compound Goes Down »
June 27, 2008
Unknown - But You Can Buy It
Posted by Derek
I sketched out a rather small molecule the other day, a perfectly reasonable looking thing, which nonetheless had absolutely no references in Chemical Abstracts. (I’d certainly like to be able to put up a drawing of the structure, but it’s something that I have a work-related interest in, so it has to stay under wraps). But it’s something with only a dozen or so heavy atoms, most of them flat and aromatic – you’d certainly expect something to have been made like it, but apparently not.
This has happened to me many times over the years. Now, you can obviously get into unknown territory immediately if you start looking for bizarre compounds: I don’t happen to have SciFinder access here on the train this morning, but I’m willing to bet that (for example) three-membered rings with one carbon, one boron, and one silicon are pretty wide open for some brave weirdo to explore. Enjoy!
But you don’t have to go that crazy to leave the paved roads behind. Many reasonable low-molecular-weight areas are only very lightly explored. You can get out of the universe of known compounds very quickly, for example, by searching for spirocycles, particularly with an oxygen or nitrogen or two scattered into the rings. Most of these would surely be interesting scaffolds for drug discovery libraries, if there were reasonable chemistry to explore them with. Even some perfectly normal looking substitution patterns of monocyclic compounds haven’t been looked into – I dreamed up a series of oxazole derivatives not long ago that no one’s ever made, and there’s nothing odd about them at all.
As you’d expect, there’s a commercial niche here. Novelty is a key requirement for patentability, so seeing no references turn up around your interesting structure is good news from an IP standpoint. (It may be bad news from a laboratory standpoint, though, because sometimes these things are unknown for a reason). But not always: there are companies that pride themselves on being able to supply such unknown scaffolds and libraries. 
The perfectly reasonable-looking diazabicyclo compound shown here, for example, has no references in SciFinder, but can be purchased on a multigram scale. (There are about fifty derivatives of that bare scaffold known in the literature, which makes it pretty much uncleared ground compared to the absolutely pulverized IP landscape around, say, piperazine). Next time you're searching for such things, refine your answer set to give only those compounds with no references, and take a look at how many of them are commercially available anyway. . .
Comments (9)
+ TrackBacks (0) | Category: Patents and IP | The Scientific Literature
POST A COMMENT
- RELATED ENTRIES
- › Pfizer's Prospects: Just Ducky
- › Happy Fourth of July
- › I Can Has Ugly Molecules?
- › More Pfizer Layoffs?
- › Leaving Comments: A Fix
- › The Gates Foundation: Dissatisfied With Results?
- › Another Alzheimer's Compound Goes Down
- › Unknown - But You Can Buy It
1. weirdo on June 27, 2008 10:01 AM writes...
It's kind of nice when the material truly is available.
But nothing is more annoying when, about half the time, it's just paper chemistry and the molecule is not actually available. So this darn company has polluted the IP waters without having the decency to make the molecule first. A particular French company is notably efficient at doing this. Many of us will never order from them, that's for sure, so it's unclear what they thought could accomplish . . . .
Permalink to Comment2. John Spevacek on June 27, 2008 10:36 AM writes...
The "Lamentations on Chemistry" blog had a post on this over the weekend. CAS did a structural search over all many simple rings and found that most of the space was unexplored which begs the question you raised: is it because no one is interested or is no one interested because it is too difficult?
Permalink to Comment3. Sili on June 27, 2008 10:39 AM writes...
I remember my supervisor being surprised that NC-CH(CONH2)2 was unknown, while (NC)2CH-CONH2 is old as the hills.
(Technically their anions. I did manage to crystallise the last one but never got round to writing it up before I fled uni. Didn't isolate the first one from the Cu complex, though. Never got around to it. I've prolly been 'scooped' by now.)
Permalink to Comment4. Jose on June 27, 2008 1:04 PM writes...
It is amazing how sparse some simple polysub benzenes are: 5-Fluororesorcinol has 18 refs and 2 preps; 3-Fluoro-5-iodoaniline has a handful and refs and a single prep.
Permalink to Comment5. andenbe on June 29, 2008 3:41 AM writes...
as for the N- respectively O-containing spirocycles you might want to check out this recent paper produced by the Carreira group in collaboration with Roche:
Spirocyclic Oxetanes: Synthesis and Properties
Permalink to CommentAngew. Chem. Int. Ed. 2008, 47, 4512.
6. Panoramix on June 29, 2008 2:33 PM writes...
There is also this weird benzoxaborole aka boron-oxygen heterocycle for the treatment of fungal infections. No the kind of ring you see daily in drugs.
Permalink to Commentdoi: 10.1021/jm0603724
7. zugor on June 30, 2008 11:42 AM writes...
Someone somewhere must have made this molecule.
Permalink to CommentTrue, this small molecule you are after may have been made, it may also be commercially available
but you don't know it. And you maybe never will.
Developing & making new building blocks & scaffold with IP potential is the easy part for the synthetic chemists. Finding the right route to market is not. Put your molecule on a commercial databases (Scifinder, Beilstein emolecules..) and it will instantly lose 90% of its value. Same for anonymous mailshots.
Too much market visibility for your new product will indeed lower its value (you lose that semi-exclusivity factor). Too little visibility and you will not be able to reach the medicinal chemists currently looking for compounds similar to yours...
8. Nathan on June 30, 2008 11:44 AM writes...
The problem is that even when you find a novel scaffold type, it's entirely possible that it has been unintentionally (or intentionally) generically covered by someone else’s IP. Then what do you do? The scaffold has not been exemplified, but it has been claimed!
Permalink to CommentAny insights? I think this would be a nice topic for a post Derek!
9. Anonymous on July 1, 2008 8:57 AM writes...
2. John Spevacek: "... is it because no one is interested or is no one interested because it is too difficult?"
Don't programs exist that assess the synthetic difficulty of molecules? Won't that tell you if it's too difficult to build? In orgo class, we heard of E.J. Corey's LHASA program and that's 40 year old!
Permalink to Comment