Skip to main content
Menu

Chemical News

Automated Route Finding (and Patent Busting)

Here’s another look at retrosynthesis software, building on the earlier Chematica paper that looked at generating new routes to known compounds. This is a more detailed look at the same idea, using the software to both analyze the existing routes to marketed drugs (and the patent landscape around them) and to come up with new ones that would slip through those existing patents.

As anyone who’s had to do this sort of analysis can tell you, a small-molecule drug can have a forest of patent filings around it. Many of these will be around the chemical matter itself (and associated things like formulations and dosage methods for it), and others will be process patents for useful ways to prepare it. Some of these may have “product by process” claims, where you’re indirectly claiming the final drug substance itself (when made by the specified route). But others will be direct claims for the process itself, not the chemical matter it produces. In many jurisdictions, the burden of proof is on the potential violator of such patents, and they have to show that they’re not infringing a patented route. Under US law, importing a drug substance that’s made outside the country via a US-patented route can constitute infringement itself (as clarified in Momenta v. Teva a few years ago). So there’s a pretty strong incentive to find new routes.

I have to say, it looks like the software does a pretty nice job of it (both the analysis and the new-route-finding). The latter is done by analyzing the existing route for the key bonds that are being formed, and then forcing the software to consider other options. Without such constraints, it tends to come up with routes that are similar or identical to the already-patented ones (which speaks well of it, actually). But if you tell it “No, I don’t want anything where that ring gets formed as part of the sequence” or the like, then it goes off-roading and comes up with something new.

For example, shown above is the program’s analysis of the patented routes to linezolid. A common theme is the synthesis of the oxazolidinone ring (and to be sure, most organic chemists would think the same way). But making those bonds “protected” makes the program come up with a different set of potential routes (shown at right). Now, this is not the most difficult synthetic problem in the world, but that looks like a perfectly creditable job of it, and the program is also spitting out literature precedents for all of these steps along with the analysis. You can certainly do this by hand, but why would you? That’s the question that such programs force us to ask, and it’s the same question that other labor-saving devices have made people ask over the years. And that’s what this and the other programs like it are: labor-saving devices. The labor in this case is not hauling laundry or digging holes, but rather the mental exertion in thinking up these pathways and the time spent in coming up with plausible procedures and precedents. The work is taking place between your ears (well, and with your fingers at the keyboard) but it’s still work.

The paper shows similar analyses for sitagliptin and panobinostat, all of which are (I would say) solid representatives of the kinds of structures and syntheses that we work with in med-chem. The former ends up being a chiral-pool synthesis rather than depending on a chiral reduction, and the latter comes up with two different approaches, depending on whether or not you allow a Pd-catalyzed coupling reaction. It would not surprise me to find that the process chemists at the various companies (both the originators and generic competition) have considered these general schemes already, of course. But they probably took a lot longer to do it!

 

27 comments on “Automated Route Finding (and Patent Busting)”

  1. Mike says:

    I would assume that the first customers of this software would be the patent departments at the pharmaceutical companies.

    1. Michael says:

      On the other hand, there might be invalidation claims based on obviousness: if the patent coincides completely with one of the options produced by publically available software, there is a chance a judge might accept putting the burden of proof on the patent holder.

      1. Anonymous says:

        Obviousness via software argument: I think that there are at least two components to the software. The “logic” based on textbookish synthesis rules but also the underlying data that directs attention from any possible route to the best routes based on precedents. Experienced chemists have read 1000s of papers and reviews and know a lot of relevant data but the computers have access to entire proprietary databases of many millions of (sometimes obscure) reactions that render some of that data non-obvious, except to computers.

        Maybe I could compare it to the heat maps generated by numerous bio screening assays. Millions of data points organized by computer analysis generate a colorful heat map that renders some key relationships obvious (to some viewers). Yeah, a key result is even obvious to a child when resolved from millions of data points into a giant red splotch on a fuzzy yellow-green-blue background but you need the (often proprietary) data as well as the logic and statistical analysis to make it so. A lot of biotech companies would lose their claims to an IP edge if “computer analysis of big data” = obviousness.

        At this time, I don’t think that the computer methods are universal enough to undermine obviousness. Maybe in the future.

  2. Anonymous says:

    It’s unlikely that a drug would be protected solely by a process of making patent. As shown above they’re too easy to circumnavigate. More helpful are compound, composition, and methods of using patents.

    1. Derek Lowe says:

      Absolutely. I see this more as a good proving ground to demonstrate the usefulness of the software, since there will have been so much work already put in to route-finding on a given compound.

    2. eyesoars says:

      It used to be that process patents lasted longer than most other patents (20 years vs. 7-14 years for others), so having one was good. I’m no longer up on the particulars (and they’re not obvious on wikipedia) so maybe they’re not substantially better in lifetime than other types anymore, since the U.S. has gone through a couple of rounds of harmonizing our patent rules with other countries’.

  3. Chrispy says:

    It’s like an arms race to cover as many bases as possible in your patents. I wonder how many of these routes will actually end up experimentally tested, though, and how many will just be prophetic nonsense.

    1. zero says:

      A patent on chemical matter should require proof of synthesis. That would severely reduce the scope of claims in certain overbroad applications.

      1. MMK says:

        It is actually the case in some jurisdictions.

        1. Len says:

          which jurisdications? any major ones?

          1. Anonymous says:

            China

          2. Len says:

            Interesting. Yes, China must count as major. What level is the burden of proof pitched at? MS ion species? NMR? I’d like to learn more, is there an online information resource you could point me to?

          3. Sisyphus says:

            India

      2. An Old Chemist says:

        According to US patent laws, if a synthesis produces as little as 0.5% of the patented compound, the patent is valid!

  4. MMK says:

    I played with Synthia (it is the name under which Chematica is being released) a bit and I must say it is as awesome as advertised 🙂

  5. Scott says:

    Let me see if I’m following this: punch in the final structure you want, and the software can give you a source-cited path to making it?

    That is really, *really* cool!

  6. old bill says:

    Software would have the advantage of avoiding our prejudices (though it may have its own) and save a lot of hard work.

    Re process patents, usual practice is surely to concentrate on protecting all (?) possible last steps, which would have to be used no matter what earlier route you devise.

    1. Anonymous says:

      “avoiding our prejudices (though it may have its own)”: The synthesis programs I am familiar with did not know a lot about rearrangements (maybe a few: Diels-Alder, Cope/Claisen, …) and certainly could not predict “unknown” rearrangements or side reactions that are sometimes more useful than the planned reaction itself.
      .
      Disclaimer: I am a human chemist and I am prejudiced in favor of human chemists.

  7. Nile says:

    So the most likely commercial application of this software will be patent-holders generating an exhaustive list of all plausible routes to synthesizing their compound, and publishing them all in a pre-emptive filing to exclude alternative ‘process’ patents.

    …And betting that the Patent office cannot check that they have tried any of them, barring the one they actually using.

    The research applications are amazing, though.

    1. Nile says:

      I’m assuming, in that answer, that the process chemists who take compounds from the laboratory to the factory are omnipotent and infallible.

      The handful I’ve spoken to are pretty damn’ close to it, within their field of expertise: but would anybody care to bet that that they have never missed a trick in every drug we synthesise, everywhere?

      I doubt that there will be a rash of slapped foreheads and the cry of “How did we miss that?” but I do expect to see a number of reviews that find a cheaper way, or cut out a difficult by-product.

      Plus, of course, the software is only as good as the data that’s in it; and every publication of a novel synthesis will now be followed, in short order, by the release of the corresponding Chematica bolt-on software module. And everybody with a compound in production – and a list of compounds that are uneconomical to synthesize today – will line up to crank the handle.

      There are real gains to all, here.

      So I wish them well with this, and I wonder what else we will see software achieving in this industry.

    2. Nile says:

      I’m assuming, in that answer, that the process chemists who take compounds from the laboratory to the factory are omnipotent and infallible.

      The handful I’ve spoken to are pretty damn’ close to it, within their field of expertise: but would anybody care to bet that that they have never missed a trick in every drug we synthesise, everywhere?

      I doubt that there will be a rash of slapped foreheads and the cry of “How did we miss that?” but I do expect to see a number of reviews that find a cheaper way, or cut out a difficult by-product.

      Plus, of course, the software is only as good as the data that’s in it; and the publication of a novel synthesis will now be followed, in short order, by the release of the corresponding Synthia bolt-on software module. And everybody with a compound in production – and a list of compounds that are uneconomical to synthesize today – will line up to crank the handle.

      There are real gains for all of us, here.

      So I wish them well with this, and I wonder what else we will see software achieving in this industry.

    3. Anonymous says:

      “exhaustive list … in a pre-emptive filing”: In the disclosure is not enabling, would it be allowed to be included in the otherwise approved patent? Even if the route is published w/o reduction to practice would that disclosure render it “obvious” forevermore?

      One recent case in the news is Berkeley – Broad CRISPR dispute. I think that Berkeley argued that once they did CRISPR in prokaryotes, it was then “obvious” to carry it out in eukaryotes (but had not yet succeeded in doing that). Broad argued that the devil is in the details and actually getting CRISPR to work in eukaryotes was patentable. (Broad prevailed in the dispute.)

      Would publishing a computer generated list of routes (with lit refs to each suggested reaction condition) undermine someone else’s effort to patent the successful synthesis? As part of the testimony, I call upon many chemists who have tried to reproduce published procedures and had to spend much time and trouble getting them to actually work on their substrate. Based on that, I’ll vote no, the computer-list should not pre-empt filing a patent on an actual reduced to practice synthesis.

      1. Anonymous says:

        It’s really not worth all this effort to preclude hostile patenting of prophetic synthetic routes. They’re still easy enough to get around. And imagine the disaster if one of them was so good it would render the existing routes permanently obsolete. Rather than potentially extending patent lifetime of your ten billion dollar molecule. You would have shot yourself in the foot.

      2. Tim says:

        “I[f] the disclosure is not enabling, would it be allowed to be included in the otherwise approved patent?”

        If your disclosure is not enabling you can still argue that the claims are enabled, for example, by common knowledge in the art. Either way you can still include in the description of the patent.

        “Even if the route is published w/o reduction to practice would that disclosure render it “obvious” forevermore? Would publishing a computer generated list of routes (with lit refs to each suggested reaction condition) undermine someone else’s effort to patent the successful synthesis?”

        Generally a publication disclosing your method is prior art, regardless if the publication reduced it to practice. There may be exceptions to a “computer generated list” depending on what you mean.

  8. Chipperwin says:

    Is chemetica available on a free trial period? Thought I’d ask on here, see whether anyone has been able to get it as such, before contacting them.

    1. M says:

      It is not publicly available. It will be released in a few months.

Leave a Reply

Your email address will not be published. Required fields are marked *

Time limit is exhausted. Please reload CAPTCHA.