Skip to Content

Drug Assays

Silicon Valley Sunglasses

The intersection between Silicon-Valley-style tech and biotech is getting a lot of attention these days. Some of it looks like it could be a productive synthesis: 23andMe hiring Richard Scheller of Genentech (update: and Robert Gentleman, today) as it starts its own efforts in drug discovery and Google bringing on Art Levinson (Genentech’s former CEO) to run its Calico subsidiary. (They recently signed a big deal with the Broad Institute and one with the University of California system). You’ll have noticed that I’m defining “productive” as “listening to people who’ve done it before”, and I make no apologies for that.
That’s because, as I’ve gone on about before, drug discovery is a very different world than hardware and software. People working in those areas have gotten used to the key capabilities of their work improving wildly and constantly getting cheaper, tendencies that have rippled out to affect everything than has a data-crunching rate-limiting step in it. And it turns out that all kinds of things do (or can be reconfigured to). But living in this environment can give a person an unusual perspective on the world.
Some of that perspective is welcome – the idea that there are always huge opportunities out there waiting for someone with enough speed and nerve to go after them, for one. That’s very Silicon Valley (and it’s also very American) and I think it’s great. But if along the way you pick up the idea that the world of apps, code, and processor speed is the default setting for the world, you can start to see everything that doesn’t advance that way as defective. That’s the Andy Grove fallacy, as I’ve called it (referenced in those links above), the idea that understanding human disease and its treatments should be pretty much like designing a new chip or writing a new app.
There’s another problem that’s not unique to the Valley, although it does tend to give people a bad case of it. That’s the “Clearly I’m smart and successful, so clearly I have something to offer in this other field over here” one. We all succumb to that one now and then; it’s human nature. You can watch Mark Cuban display it here, with respect to medical testing.
But here are a couple of recent examples of the more localized problem. I wrote last year about Emerald Therapeutics, an outsourced-lab-assay company backed by Peter Thiel (who may also be interested in their antiviral therapy ideas). Here’s another article on them, and it asks, in so many words, “Why is new drug development so comparatively torpid when app development is so torrid?”. I couldn’t provide a more succinct version of the Silicon Valley/biopharma disconnect if I tried.
According the article, the folks at Emerald “. . .think it comes down to the difficulty of running experiments in the life sciences”. But I’d like to propose that this difficulty, at least for early-stage work like Emerald is proposing to do for people, is largely a matter of contrast. If you’re used to being able to sit down and bang out code, any time, anywhere, with all kinds of tools (libraries, compilers, virtual machines, what have you) at your fingertips, then yeah, working up a new assay protocol in a cell line is going to seem agonizingly slow. Multibillion dollar ideas can be cranked out in the coding world very quickly, if you hit the right place at the right time, but just you try that in the lab. Now, I have no problem with Emerald running assays for people, although it may yet be harder than they’re thinking. But they’re not removing as much of a bottleneck as they might think. The real bottlenecks are figuring out what assay to run, and what to do with the data once you have it. Can’t outsource those.
Then we have this piece on genomics at TechCrunch. Experienced readers will get a very 1999 vibe from it; it’s full of the wonders-of-the-genome stuff that was so thick on the ground back then. (That makes it simultaneously cute and annoying to see it all again, of course). “One App Away” is a headline from the article, which is suffused with the next-big-thing attitude that you’d find from someone who’s figured out how to social-gameify the process of splitting the check at a bar, complete with a three-second clip of everyone hoisting drinks with the amount they chipped in hovering over their heads. That probably already exists.
But you know, I don’t really disagree that much with many of the conclusions in the genomics piece – just the pace at which things will happen. That’s what I think has been Valley-ized there, the idea that very, very soon now something will just wildly, exponentially take off. As much as I might like to see something like that happening in biopharma, though, I can’t quite make myself believe it. Technology, Silicon Valley style technology, is human-designed and human-optimized for other humans. As human beings, we’re playing on our home turf there. But the biology of disease is an away game if there ever was one. The inner workings of cells and the ways that they work together are flat-out alien compared to anything we’ve ever built ourselves. People who are used to coding up apps have never experienced anything like it, and many of them don’t seem to realize that they haven’t. Expecting the sorts of behavior that you get from human-built technologies, and expecting the same effects from the techniques that work to optimize them, is an expensive accident waiting to happen.

37 comments on “Silicon Valley Sunglasses”

  1. Hap says:

    “One App Away” is a headline from the article, which is suffused with the next-big-thing attitude that you’d find from someone who’s figured out how to social-gameify the process of splitting the check at a bar, complete with a three-second clip of everyone hoisting drinks with the amount they chipped in hovering over their heads.

    I’m not the best exponent for a classical/humanities education, but this seems like an argument for it in a nutshell – the idea that apps and computers can make problems of human interaction disappear. People are mean and selfish for lots of reasons, and we haven’t figured out how not to be so (or even if we should not be so). The presumption that people will behave properly given the proper app is either frightening (if it can be done) or hubristic (if it can’t). And humans, unlike our biology, can be reasoned with (sometimes).
    If someone wants to spend their money to figure out biology and improve people’s lives, there no reason to complain about that. I just wish it didn’t come with the assumption that the rules of the world of technology (which hasn’t long existed) are those of the world at large, and should be.

  2. Todd says:

    Perhaps it’s my assay development roots here, but I don’t think Peter Thiel’s assay shop is complete foolishness. When it comes to good statistics and DoE, the more power the better. It doesn’t eliminate the biologists of course, but give me an assay that’s more robust and even fails more predictably, and you’re definitely going to move the needle. It won’t be an revolution, but it have some very strong evolutionary effects.
    There’s a lot of statistics that isn’t done as well as it should, and more computer power will help.

  3. Derek Lowe says:

    #2 Todd – I don’t have problem with the assay shop, either, although there are plenty of other places that will run assays for you (albeit in a less flexible sense, if I’m getting Emerald’s business model right). Setting up some of these off-site is going to be trickier than it might look, but I still think that they should have at it. But the idea that this is going to be some sort of missing link of technology, that’s what I object to. It’ll help; it won’t solve.

  4. steve says:

    I think they’ll have an enormous impact not only on therapeutics but on diagnostics as well, which is probably the low-hanging fruit. Look at Google’s contact lens for diabetics, Apple Research Kit, etc. Hell, if the software guys can just improve compliance they’d make huge inroads into improving our health care. In terms of therapeutics, new chemical entities aren’t the only game in town. Being able to tailor therapies to genomic variation could also make a big impact. Being able to mine large datasets to identify potential drug repurposing is another. One could go on and on – predictive apps that can track infectious disease variation in real time so the next flu vaccine doesn’t miss the mark like this year’s. There’s often a lot of cynicism and self-justification in these discussions; enough people suffer from horrid diseases that if silicon valley wants to apply their skills and make a difference more power to them.

  5. John Wayne says:

    @4 Steve, I think the idea of optimizing doctor/patient interactions with Apps is a great idea, and one that will probably improve health care quite a bit. I’d download and use an App to help remind me to take a pill on a regular schedule, etc. If a Doctor recommended an App, I would be very likely to try it.
    I think the skepticism (which is justified) is about filtering large data sets for undiscovered breakthroughs; it could happen, but doesn’t seem likely. What would be great is a better tool (or set of tools) that allows a researcher to ask flexible questions (run computational experiments) on their data. This could be a big deal.
    The caveat here is that lots of smart people work on this every day, and it isn’t easy. When you read these articles they have this breathless tone that nobody has bothered to use computers in drug discovery. Not only is that polarizing, it is also insulting to the experts. If you want to help, great; the more eyes the better. Don’t let your publicity and marketing strategy isolate you from the people who are already working in the field; it isn’t going to help you or patients.

  6. Geb says:

    If you really want to get the techie crowd to understand why medicine and biology are so difficult, explain debugging the human body by software analogy.
    Suppose there was a neural-network database engine, that would query its records according to some arcane undocumented (and indeed undocumentable) trained-in process. The database layout, indexing, data structure and so on were all designed by the neural network, with no attempt to influence it by externally applied human sanity. Nobody really knows what data corresponds to what, except in extremely vague empirically determined terms. Input and data retrieval is achieved by flashing different printed photos of dogs in front of the server webcam. The QA department refuse to give any feedback other than “Sometimes it works.” Nobody knows how many dev teams have worked on this over the years, or recorded what features they (attempted) adding.
    You are now tasked with debugging a problem in which duplicate data is sometimes lost.
    Any variants on the phrase “abandon the whole thing and rewrite it from scratch” are an instant fail condition.
    Begin.

  7. Anonymous says:

    @Hap “The presumption that people will behave properly given the proper app is either frightening (if it can be done) or hubristic (if it can’t)”
    I think you are getting carried away there. Splitting a check at a bar/restaurant amongst friends is usually more of a math problem (that an application can easily solve) or a race to quickly grab the check. Maybe you need some new friends if they aren’t willing to at least pay their share.

  8. Esteban says:

    The unbridled enthusiasm/evangelism of the Silicon Valley tycoons is both indispensable for technological progress and lampoon-worthy at the same time. The display of hubris can be jaw dropping at times. Yes boys, please show us how easy drug discovery can be if we were only as clever as you.

  9. Mike says:

    @5
    I would even take it further. I think the most important contribution technology can make is an app/interface that integrates diagnostics (which obviously have room for improvement), healthcare data, and video-conferencing services that facilitate doctor-patient interactions, while also reinforcing compliance.
    A huge amount of medical costs come from preventable diseases. Granted, some people will just make terrible decisions on how to treat their body, but the above kind of technology could improve the access and quality of routine medical care and treatment of chronic conditions. Whether or not people would actually change their behavior in response to this sort of system remains to be seen, but I think it’s worth investigating.
    I think that’s the arena where technology will be most successful. The problem is much better defined, easier to tackle, and probably has even better overall value (though likely not monetary) than drug discovery, which can be far more lucrative.

  10. steve says:

    In 1835, Auguste Comte stated that humans would never be able to understand the chemical composition of stars. After all, the sun is so hot there is no way you could ever get a sample to test. Then spectroscopy was invented. The people who always end up looking foolish are the ones who say something can’t be done, however logical the reason. Bringing enormous computational skills to drug discovery can only help. Being naive at the beginning is usually a big plus for scientific advancement because you’re not inhibited by what us old folk say is impossible.

  11. Todd says:

    #3 – Perhaps it was my grad school exposure to computational biology students, but I’m a bit more sanguine about explaining biology problems to IT types. Once you know how to bridge the language gap, you can focus them in the right direction to problems that are legitimately useful. Use the words “fuzzy continuous system”, and suddenly they both get the challenges and where they can be useful.
    Also, the update about the founder of the Bioconductor project joining up with 23AndMe shows that they aren’t joking around. This is someone that knows what they are doing on both sides of the gap.

  12. steve says:

    I’ll give you another example. When I was a graduate student in immunology, there was a huge debate about whether a particular immune region of HLA, called “I-J” existed, It was defined by some antibodies but the genetics was really weird and no one could immunoprecipitate it. Then I attended a seminar by Lee Hood who blew the entire audience away with the advances he was making in protein and gene sequencing (first time I ever heard anyone working in femtomoles). People jumped up to ask questions. Before they did, he said as an aside, “by the way, we’ve sequenced the entire HLA region and I-J doesn’t exist”. A new technology everyone was skeptical about – newfangled DNA sequencing – blew away one of the biggest controversies in immunology, one that couldn’t be settled with the tools available at the time. The idea that silicon valley can’t do something like that for drug discovery may be of comfort to us who toil in the lab with beakers and pipettes but one day we may look pretty old-fashioned.

  13. adam says:

    The problem with Silicon Valley types is that once they come up with the “next big thing”, they tend to force it on everyone with any even somewhat-related problem (see also, how little you can do on the inter webs without a Google account). When applied to drug discovery, that attitude will likely be quite dangerous.

  14. dr z says:

    the entire tone of this asinine inc.com article is set with the opening sentence comparing a food-delivery app with a new antibiotic. seriously?
    i have to say, my fav aspect of the inc.com article is the fact that if you scroll down to the next article, it’s about the inventor of the pet rock who’s being heralded as the greatest entrepreneur of all time…

  15. steve says:

    #13 – A quote from Steve Jobs: :The over-all point is that new technology will not necessarily replace old technology, but it will date it. By definition. Eventually, it will replace it. But it’s like people who had black-and-white TVs when color came out. They eventually decided whether or not the new technology was worth the investment.”
    Or, better yet, Steve Brand: “Once a new technology rolls over you, if you’re not part of the steamroller, you’re part of the road.”

  16. steve says:

    Woops, sorry, I meant Stewart Brand

  17. Esteban says:

    #10 said: Bringing enormous computational skills to drug discovery can only help.
    Drug companies have probably already spent billions in such an effort, so it’s not like no one is trying. And with billions in potential revenues at stake, it’s not like the incentives are misaligned. Can the SV crowd do it better? Maybe, though I doubt it. Still, as long as it’s private money being spent, I’ve no objections, I’m just lampooning the hubris, that’s all.

  18. steve says:

    #17, Maybe spending billions on marketing to defend your existing drugs isn’t the same as bringing in a fresh perspective and capability. I lived in Massachusetts during the meltdown when no one thought that PC’s would replace mainframes. When was the last time you bought your computer from Wang?

  19. pete says:

    “Silicon Valley Sunglasses”: I speak as a biologist who lives in The Valley and has had a bit of experience with in silico approaches to drug assessment following lots of experience in bench-based discovery.
    I think there’s change for the better among the more serious-minded Computer Science types who turn toward Drug Discovery (DD) questions these days. So where there’s still evidence of “Silicon Valley Sunglasses” in some of business plans of startups engaged in DD, I think more appreciate the shortcomings & intrinsic hubris of the Andy Grove-ian/Ray Kurzweil-ian pronouncements about cracking human biology.
    DD is an extreme case of an “Away Game” (great analogy, Derek). We didn’t design it; we’re the blind man inside a strange dark castle trying to lay hands on a golden key.

  20. anon says:

    wow steve sure feel strongly… clearly since there have been progress in the past, the future must be written out exactly as the silicon valley people are saying it will. because when have they ever not followed up perfectly with every “forward looking statement”
    also your interaction with #17 in #18 is just a straight up attack and completely avoiding his (civil) point.
    dont be so rude.

  21. unchimiste says:

    #10, your point about the composition of stars is valid. But it’s also counterbalanced by a even more famous Auguste Comte’s theory: the Theory of Science. According to this theory, problems in biology are more complicated to solve than problem in math. That what the people writing apps (which is conceptually doing math) for a living don’t understand. The problem is biology is the fragility of the theoretical background, it’s hard to reduce biology into equations.

  22. steve says:

    #20, I in no way was “attacking” #17. I just was pointing out that large pharma spending billions on something is not a way to judge the validity of a new approach. I have no idea if silicon valley is going to cure cancer or not. In general, I find the board here, while often enlightening, to be very cynical about any new idea. These guys want to spend lots of their money and time trying to fight disease. Maybe they’re over-exuberant. Maybe some will succeed, maybe others won’t. But all innovation starts out that way – lots of old school people saying something will never work and a few true believers championing another way proving them wrong. Will that be the case here? Only time will tell but I wouldn’t be so sure that the way we’re currently doing things is the only way that can work.

  23. Pedantic Spaker says:

    “The real bottlenecks are figuring out what assay to run, and what to do with the data once you have it. Can’t outsource those.”
    Not to dispute the substance of your argument, it is poorly worded: An inability to outsource thinking is not the problem.
    Paying universities to think about specific questions, if not the entire purpose of the NIH (on the academic side of the field), is not too far from what it actually does.
    And on the industry side, companies routinely buy the rights to compounds from other companies, and as far as I can tell, that necessarily involves buying the “figuring out” that was involved in their development.
    The real problem is that thinking is far more difficult than calculating or collecting data, and is hence more expensive and unreliable.

  24. Esteban says:

    steve: The fact that billions are spent on drug marketing is a non sequitur. I’m sure Oracle/Apple/Microsoft spend a decent amount on marketing – that doesn’t mean they are being short-sighted with their R&D spend. Apples and oranges. Fresh perspective is fine and like I said, if it’s private money, have at it. Don’t be a sucker for the hype though – there are a lot of smart and motivated scientists in the drug business and drug discovery is not as tractable as the Sand Hill Road billionaires think it is.

  25. steve says:

    Yeah, the marketing comment may have been a bit snarky, my apologies. It just reflects how pharma has shifted its priorities over the years. Yes there are a lot of smart people in pharma but as this board has reflected many times, a lot of them have been fired and R&D has suffered due to emphasis on marketing rather than new product development. Having spent my life in startup biotechs, I can tell you that without the hype these guys won’t get the funding. No VC is going to fund a startup that says, “drug discovery is hard, we don’t know if we can succeed applying our computer expertise to it since biology is complicated but gosh darn it we’re going to try!”. They will, however, fund someone with a track record in computers who says they can apply their expertise to drug discovery and have an impact. I’m always in favor in taking a fresh look at a problem even if all the experts in the field say the old way is better. Guess that’s why I’ve never been attracted to large pharma, just to small, entrepreneurial startups. Different strokes…

  26. John Wayne says:

    @Steve. Hey, we’re not even saying that you are wrong about the proportion of money going to marketing vs. research. The whole things feels a bit dirty to even big pharma scientists, but it is necessary (probably). I’ll paraphrase your comment about hype: everybody roots for the underdog, but nobody bets on him. If you want people to buy in, you have to tell a compelling story.
    I’m a big fan of startups myself, but I’m glad I did my time in big pharma to learn their ways of doing things; it’s another tool in the belt. In the 80’s, I remember that most of my friend’s parents worked at Digital; now, I’m not sure if the company exists. Some people assumed that nothing would change, and they weren’t right.

  27. Esteban says:

    Fair enough, the marketing side of the industry gives me the icks too, but I never see those people – I just see the same ads on TV that everyone else does. Allowing pharmas to advertise direct to consumer is what drives the ad spend. Perhaps that means less for R&D, though that’s hard to know for sure. Yes, plenty of good people have been let go, but there are still plenty of good ones left. I suppose hype comes with the territory with start ups, but the condescending discussion of doing it better rubs a lot of us the wrong way.

  28. oldnuke says:

    @26 Digital Equipment bought by Compaq bought by Hewlett-Packard
    After going through all of those corporate digestive tracts, all that remains is cr@p. Unfortunately, they (DEC) had great engineering and lousy marketing.
    I’m not sure if HP has any greatness left in it. Certainly none in the executive suite. Anymore, they are printers, expensive ink and me-too computers.

  29. There’s a bigger problem here than you think. Virtually all of the Next Big Thing software isn’t even hard. Most of these guys don’t have a clue what hard problems look like.
    Next big software things are all in the idea. The implementations are generally easy. If there’s any real hard engineering it’s on datacenter side.
    These people, as near as I can tell, don’t actually know this. They think they’re solving hugely difficult problems when in fact they’re getting lucky with seemingly crazy ideas that are easy. The fact that they hire 5000 bozos to constantly move the buttons around in the UI is not evidence that they’re solving hard problems.

  30. Andrew Dalke says:

    I think Autodesk’s involvement with HyperChem back in 1990 should be a cautionary tale. One of the Autodesk founders describes why they should be involved in molecular modeling, at https://www.fourmilab.ch/autofile/www/chapter2_82.html . Some choice phrases:
    “The safest bet in the world is that computing power will continue to grow at an exponential rate while costs stay constant or fall.” … “Molecular modeling has the capacity, like CAD, to soak up all the additional computing power anticipated for the next decade, and in the process, expand the market for computational chemistry just as AutoCAD has done for CAD.” … “Every step we take, and every success we have in this developmental period will just put us in a better position for the time when molecular engineering explodes into exponential growth in the manner integrated circuits and microprocessors did in the 1970s and 1980s. And if it never happens, then we’ll still be at the centre of the rapidly growing biotechnology market, where molecular modeling is already essential.”

  31. anon says:

    Steve: as Derek often points out, Pharma spends a much higher percentage on R&D than nearly all other sectors, so attacking their money on marketing is not only a non-sequitor, its also misleading, because they people you are rooting for are probably far worse in the issue you are decrying.

  32. Dan Munro says:

    Great follow up to a piece that Steve Blank wrote about 3 years ago: Why Facebook is Killing Silicon Valley.
    Key quote:
    “If investors have a choice of investing in a blockbuster cancer drug that will pay them nothing for fifteen years or a social media application that can go big in a few years, which do you think they’re going to pick? If you’re a VC firm, you’re phasing out your life science division. As investors funding clean tech watch the Chinese dump cheap solar cells in the U.S. and put U.S. startups out of business, do you think they’re going to continue to fund solar? And as Clean Tech VC’s have painfully learned, trying to scale Clean Tech past demonstration plants to industrial scale takes capital and time past the resources of venture capital. A new car company? It takes at least a decade and needs at least a billion dollars. Compared to IOS/Android apps, all that other stuff is hard and the returns take forever.”
    Point being – we’re losing our moxy (and interest) to solve hard problems. Easy ones are just too lucrative.

  33. Mark Kaganovich says:

    Derek, Great article! I wrote the Techcrunch piece you cited. Thanks for checking it out. I did want to clear up a few things – I don’t really delve into the timeline of when genomics will go mainstream and be applicable to healthy people. You mention your disagreement with the timeline, so I’m wondering where that comes from. My point in the article is that genomics has strong network effects and that while it is largely irrelevant to many people now, it has all the ingredients of an exponential growth industry. Happy to hear thoughts on that, and perhaps others disagree, but I also want to point out that while your idea for a group payment app sounds nice, my background is not at all in consumer web (I have an AB in Biochem and Computer Science and a PhD in Genetics and have always been a scientist). Same goes for the others at the company. I’m excited to hear people with more drug discovery experience than myself chime in on how computers can help (although our company doesn’t do drug discovery, but perhaps biotech experience in general), but disparaging “Silicon Valley” engineers as mere marketers too young or too dumb to build “real technology” is strange, insecure, somewhat irrelevant, and naive. We all know software has a role to play, and Silicon Valley-style startups are good at making software. It’s true that many technologists that don’t know basic biology extrapolate trends too readily and predict immortality somewhat too eagerly, but many are quite smart and motivated to expend great effort to solve hard problems. In fact, at our company, SolveBio, we DO know biology, quite well, and are motivated by the complexity of medicine. I wasn’t in the industry in ’99, and I’m sorry if our excitement is an annoying reminder of the bright-eyed late 90’s, but we’re developing technology that is new and real, and progress depends on it – otherwise every year from now on will sound like ’99.

  34. ab says:

    @ #32 Steve-
    Just looked at the last graph in the link you shared. According to that graph, there are 13 companies that spend more than 10% of revenues on R&D: 3 are “technology” companies, 10 are pharmaceutical companies. Of the 7 companies that spend more than 15% of revenues on R&D, 1 is in technology, the other 6 are pharmaceutical. These data strongly discredit your viewpoint.

Comments are closed.