Skip to main content

Chemical News

Med-Chem, Automated?

See what you think about this PDF: Cyclofluidics is advertising the “Robot Medicinal Chemist”. It’s an integrated microfluidics synthesis platform, assay/screening module, with software to decide what the next round of analogs should be:

Potential lead molecules are synthesised, purified and screened in fast serial mode, incorporating activity data from each compound as it is generated before selecting the next compound to make.
To ensure data quality, each compound is purified by integrated high pressure liquid chromatography (HPLC), its identity confirmed by mass spectrometry and the
concentration entering the assay determined in real time by evaporative light scattering detection (ELSD). The compound’s IC50 is then measured in an on-line biochemical assay and this result fed into the design software before the algorithm selects the next compound to make – thus generating structure-activity relationship data. The system is designed to use interchangeable design algorithms, assay formats and chemistries and at any stage a medicinal chemist can intervene in order to adjust the design strategy.

I can see where this might work, but only in special cases. The chemistry part would seem to require a “core with substituents” approach, where a common intermediate gets various things hung off of it. (That’s how a lot of medicinal chemistry gets done anyway). Flow chemistry has improved to where many reactions would be possible, but each new reaction type would have to be optimized a bit before you turned the machine loose, I’d think.
The assay part is more problematic. There are assays suitable for in-line evaluation like this, but there are plenty of others that aren’t. (I would think that SPR would be particularly well-suited, since it operates in flow, anyway). But that prior optimization that the chemistry needs is needed even more here, to make sure that things are robust enough that the machine doesn’t generate crappy numbers (and more swiftly than you could do by hand!)
The software is the part I’m really wondering about. How is this thing picking the next round of analogs? Physiochemical descriptions like logD? Some sort of expert system with med-chem lore in it? Does it do any modeling or conformational analysis? Inquiring minds want to know. And I’d also like to know if they’ve sold any of these systems so far, and to hear some comments from their users.
Update: here’s one.

25 comments on “Med-Chem, Automated?”

  1. Anonymous says:

    I wonder if a compound “discovered” this way would be patentable…

  2. Henry's cat says:

    Chemists destroying Chemistry. Marvellous. We’re having a hard enough time existing at the whim of accountant CEOs and now we have ex industry scientists and a Cambridge professor trying to knock a few more nails in our coffin.

  3. Anonymous says:

    And between runs it will vacuum the office and make you coffee. I’ll take two!

  4. Justin says:

    You don’t need to be a chemist any more, just a plumber and electrician…for when the thing breaks down.
    And #1: Yes, first-to-file

  5. Lyle Langley says:

    I’m a medicinal chemist that can do the robot…does that count as the same thing?

  6. Anonymous says:

    @5 Brilliant!
    Well, I guess we are done for. The robot is no doubt powered by a computer, and we all know how good computers are at discovering drugs. Also, good thing a correct mass is always 100% confirmatory for compound identification. I am thinking one would spend more time sorting through the questionable data this thing would generate than actually doing the chemistry manually.

  7. DCRogers says:

    The abstract of the linked article: “Drug discovery faces economic and scientific imperatives to deliver lead molecules rapidly and efficiently. Using traditional paradigms the molecular design, synthesis, and screening loops enforce a significant time delay leading to inefficient use of data in the iterative molecular design process. Here, we report the application of a flow technology platform integrating the key elements of structure–activity relationship (SAR) generation to the discovery of novel Abl kinase inhibitors. The platform utilizes flow chemistry for rapid in-line synthesis, automated purification, and analysis coupled with bioassay. The combination of activity prediction using Random-Forest regression with chemical space sampling algorithms allows the construction of an activity model that refines itself after every iteration of synthesis and biological result. Within just 21 compounds, the automated process identified a novel template and hinge binding motif with pIC50 > 8 against Abl kinase — both wild type and clinically relevant mutants. Integrated microfluidic synthesis and screening coupled with machine learning design have the potential to greatly reduce the time and cost of drug discovery within the hit-to-lead and lead optimization phases.”
    My knowledge of some of the authors suggests that the modeling method is likely to be a reasonable choice – they’ve been in the field long enough to know the pitfalls of overconfidence.
    But in any case, if you’re mistrustful of the software, just wire-in your human med chemist in its place, and the remainder of the work still seems interesting.

  8. Anonymous says:

    Thank you very much, Mr. Roboto
    For doing the jobs that nobody wants to
    And thank you very much, Mr. Roboto
    For helping me escape just when I needed to
    Thank you-thank you, thank you
    I want to thank you, please, thank you
    The problem’s plain to see: too much technology
    Machines to save our lives. Machines dehumanize.

  9. Becks says:

    Have you noticed you seem to not like any new med chem approaches or strategies that were developed post-1994? Just an observation.

  10. anon the II says:

    I’m gonna call serious BS on this whole thing. I don’t doubt that you can cobble something together and sitting by with string and baling wire you can get enough through to write a specious though publishable paper. The whole problem with flow systems is precipitation. And it’s generally catastrophic to any ongoing process. It’s enough of a challenge just running a reliable mass guided purification operation. So when you string essentially all types of chemistry, with integrated purification and then assaying, you’re probably gonna have to take the system apart for about every 3 compounds made.
    Some poor jerk is probably, as we speak, trying to get all the copper salts out of the system so they can run the next publication. Maybe they could get Becks to do it.

  11. JC says:

    I for one look forward to our Robot Synthesis Overlords.

  12. exGlaxoid says:

    I have to second #10. Another big pharma company spent well over $200 million building 2 automated chemistry buildings and filling them with automation that was going to automate all of synthesis, purification, handling and screening of pharmaceuticals.
    The systems clogged constantly and broke often, the software was too complex to anyone to use, and the chemistry was very limited and made amide and triazines. The company also bought two other companies that automated chemistry and made vast libraries of amides and triazines, both of which eventually also went away. The real issue is not finding leads, it is discovering drugs.
    Both buildings are now empty and each one never produced as many compounds as a small team of chemists did at each of two other small groups which were not given the budget to fully automate chemistry, but merely allowed to buy some useful tools for simple automation, some of which worked really well, like automated vial weighers, prep HPLC systems, and Gilson 215 liquid handlers. And several other groups of chemists made simple compounds the old fashioned ways, which lead to more more clinical drugs than either.
    So of course all of the groups of chemists were laid off, and the few groups than made zero marketed drugs were kept. I guess that statistically speaking, they are due.

  13. Hap says:

    @9: Have any of them actually worked at finding drugs? Spending lots of money and making management and investors ooh and ahh doesn’t count, and unemploying much of the drug industry doesn’t count either.
    If someone can get something to do what a med chemist does much cheaper and actually have it work (be able to find drugs), then (almost) everyone will be better for it. Most of the “advances” have been made to sell papers, equipment, or management pillaging to unwary investors, but haven’t led to actual drugs. I don’t know if this is one of those, because I sort of like flow chemistry, but when the hype machine spools up, it doesn’t raise my hopes.

  14. Lyle Langley says:

    Thank you JC, or should I say Kent Brockman. We are being taken over, “conquered” if you will. I’d like to remind them that as a trusted human medicinal chemist, I could be helpful in rounding up others to toil in their underground sugar caves.

  15. Hap says:

    @12: I think someone made this point in Men at Work: a .240 hitter hitting horribly probably isn’t due for anything good; unless he figures somthing out, his output probably looks like what it should. A career .310 hitter hitting .240 or having a bad streak probably is due (unless he’s come off steroids or is permanently injured); you could reasonably expect hitting streaks from him, or at least more than a bad streak suggests.
    I thought pharma management was great with statistics.

  16. David Formerly Known as a Chemist says:

    This reminds me of the “Charlie and the Chocolate Factory” movie, where Charlie’s father was laid off from his job putting caps on toothpaste tubes because he was replaced by a machine, only to later be hired back to repair the machine. Ex-chemists like myself will someday unclog these flow machines, scrape out the spent palladium catalysts, flush the system with DMSO to remove all the precipitates, and bore out the polymers with a pipe cleaner. All those years of grad school and postdoc will pay off!

  17. Dr Manhattan says:

    I must confess. I have been secretly working on “The Robotic CEO” for several years. It too has an Abstract, and it is so similar to the one above, I think I was ripped off!:
    Potential leaders are synthesized and screened in fast serial mode by HR, incorporating activity from each McKinsey Group and multiple PowerPoint decks generated before selecting the next wrong decision to make.
    To ensure CEO bonus quality, each decision made by the Robotic CEO is filtered through layers of multiple committees, diluting the responsibility and integrating high pressure goals (HPG), to insure individuals lose their identities, confirmed by mass layoffs and the concentration of pay at higher levels.
    A light scattering of 2% bonuses (LSB) insures enough workers to continue the company & CEO employment. The company’s ROI is then measured in an on-line Investment Banker call and this result fed into software before the algorithm selects the next CEO bonus– thus generating structured activity in the CEO bank account.

  18. Derek Lowe says:

    #9 Becks – well, it’s not like I’m wild about the pre-1994 methods either. But even though there are new ideas I find interesting and worthwhile, I just don’t see any of them coming in and totally changing the world as we know it. I’ve done plenty of automated flow chemistry, for example, but I wouldn’t like to claim that it’s going to fix drug discovery.

  19. Anonymous says:

    Getting to a drug requires more than simple SAR analysis. Chemistry evolves as a more profound understanding of the target develops later in LO. There are too many simpletons who assume the art is just changing R groups around the same core. The bottleneck is not squeezing reagents into the’s taking time to do the right thing. The numbers game doesn’t work and making it robotic is optimising failure. There will be small gains and it’s good to go from synthesis to biological readout in the blink of an eye but not if you take decisions by proxy. Chemists are complex beasts who understand the importance of hedging their hypotheses and not just using algorithms to chase actives but actively trying to break their own hypotheses. Good luck. The future is bright. Let’s just buy a drug printer.

  20. anonymous says:

    My former (big pharma) company invested heavily, both time and dollars, in an automated synthesis capability. At the level of technology it was extremely impressive and could be programmed to deliver on a much broader variety of chemistries than other systems that I’d observed. For subsets of SAR, such as early hit expansion, it could be made to function reasonably well and in its ‘sweet spot’ could be incredibly productive. However, as several of you have recognized, the challenge came when you needed to evolve your lead based upon the overall optimization parameters. Inevitably, the conflict arose between the chemistry that could be done with the automation and the chemistry that needed to be done. Without the ability to break away from the limitations of the automation the project was doomed to compromise, making an already difficult endeavor even more so.

  21. Anonymous says:

    Lol ExGlaxoid, isn’t it so that the big pharma company in question also jetissoned it’s tech. dev. department, which then went off, got venture capital and started a company in Welwyn using flow technology to automate med. chem. ….. I can’t think what they called that company …. what was it again…..

  22. Doug Steinman says:

    Mass spec is not proof of compound identity. It does not differentiate between cis and trans isomers or ortho, meta para aromatic substitution, among other things and all of those structural variations can have significant effects on biological activity. I would challenge any patent that does not have NMR data for the claimed compounds.

  23. Nick K says:

    #17: Comment of the week, Dr Manhattan!

  24. Wage_Slave says:

    Did anyone else look at the experimental in the paper linked in the ‘here’s one’ above? It makes interesting reading; I am not sure that they actually used their flow robot system to generate the results at all!

  25. BL says:

    Agree with the skeptics. By trying to integrate the full process, from synthesis to assay, on one chip (or equivalent) you force a series of restrictions that make the attempt useless. You are restricted in your choice of chemistries, in your purification and analysis methods, and in your assays. Catalytic antibodies, anyone?

Comments are closed.