Verge Genomics (YC S15) Wants to Cure Neurodegenerative Diseases with Algorithms | Hacker News Reader

Verge Genomics (YC S15) Wants to Cure Neurodegenerative Diseases with Algorithms

techcrunch.com

52 pointskatm55 comments

The breathless tone of this TechCrunch article reminds me of this commentary on Atomwise (also a YC biotech): http://pipeline.corante.com/archives/2015/03/26/tone_it_down...

In my opinion, YC has no idea what they're doing in biotech. Here's a much more credible neuro disease startup that was recently launched: http://www.forbes.com/sites/matthewherper/2015/05/14/former-...

I'm sure Alice and Jason are very intelligent and motivated, but the smart money's on Marc Tessier-Lavigne any day of the week. The industry dynamics of biotech are totally different - the small and nimble upstart with unconventional thinking rarely wins.

This is what happens with media coverage of pharmaceutical development, and often science in general. Journalists want stories that people can relate to, that don't have the uncertainty that is inherent in science, and that will grab the public's attention. The ridiculous tone is often something that horrifies the people being reported on, but seems very difficult to escape.

It doesn't help that here, I think, we're seeing a community and media largely centered around software development trying to deal with a field that is far, far riskier, and has far less certainty at every step. Good ideas in software might not catch on, but good ideas in science more often than not turn out to be entirely wrong. Do a closed beta of your software, too, and while your users might not end up liking the software much, it's very unlikely that it's just going to fundamentally fail to work at all. Clinical trials can often end up that way.

Whether YC has an idea on biotech or not, I'm very happy to see more startups in this domain. I think we've had enough social networks and mobile app frameworks. Seriously cool that YC starts backing companies that want to bring real change to the world.

Mathew Herper's hype is more subtle than Tech Crunch's but it's hype nonetheless. He plays his favorites just like Tech Crunch. At least he's not as unseemly as Tech Crunch though.

The biotech companies in the last batch were mostly reasonable, if a little too breathlessly overhyped by their press launches. Say what you will about that Atomwise article, they're playing in a well-known space. There are existing companies (generally small) selling exactly that sort of service to drug companies. It's an open question whether or not their particular approach is sufficiently better to give them a sustainable advantage in what is an extremely competitive, low-budget space (relative to med-tech), but the business model is plausible.

Most of the biotech companies coming out of YC are what I'd characterize as total fliers -- unlike the company highlighted by the Forbes article, these startups aren't being founded by a collection of gray-haired biotech executives with a collective 200 years of research experience in a single disease. And that's fine. One of the biggest problems in biotech right now is that everything is too stodgy and conservative. When I got out of grad school, there was no way of starting a biotech company without putting in 20+ years in academia or at drug company. You wouldn't get funding.

If funding a few grad students for oddball biotech startups leads to a single success, it has the possibility of really opening up the field. And in the worst case, very little money will be lost. I think it's a good thing. If nothing else, maybe it'll turn the attention of smart, ambitious people to something more important and meaningful than photo-sharing apps and YuppieFoodDelivery.com v87.0.

Let a million flowers bloom...as long as they all don't try to crowd into San Francisco, that is.

The thing is this kind of idea is not new. Finding new targets is hard science that needs a ton of experiments to validate that your theoretical models even work. For now this is low risk and no cost to the pharma partner, but it's a not a new model and in my experience not one that works.

"The thing is this kind of idea is not new."

...most startup companies aren't doing anything new. How many different food-delivery services has YC funded? Photo-sharing services? Video startups? AirBnB wasn't anything new in 2007, either. Dropbox was criticized because everyone knew that file sharing had been done before.

This goes back to what I was saying about biotech being stodgy: the one thing the software industry has going for it is that it's willing to fling seed capital at re-tread ideas. If success is mostly about being in the right place for luck to occur, then the secret is putting enough companies in a place to experience luck. I think that's basically what YC does.

So-called "biotech investors" tend to want to wait around for commercially ready tech to just fall out of academic labs, so that they can avoid putting any investment at all into R&D. There are probably a lot of inexperienced teams that could make a business in biotech if only they had access to cash.

> ...most startup companies aren't doing anything new. How many different food-delivery services has YC funded? Photo-sharing services? Video startups? AirBnB wasn't anything new in 2007, either. Dropbox was criticized because everyone knew that file sharing had been done before.

There's a fundamental difference between trying something again where the customers weren't receptive and trying something again where the product couldn't be created. Customer reception can change over time, often quite rapidly and drastically, making previously failed business models successful (ex: Webvan -> Instacart). But nature doesn't just change (at least not over short timescales).

Moreover, as cge explains elsewhere in this thread, you cannot really iterate in biotech:

> Good ideas in software might not catch on, but good ideas in science more often than not turn out to be entirely wrong. Do a closed beta of your software, too, and while your users might not end up liking the software much, it's very unlikely that it's just going to fundamentally fail to work at all. Clinical trials can often end up that way.

Contrast this to business model-based innovation in tech, where a slight tweak can be the difference between a unicorn and a dud.

The difference between success and failure in biotech lead generation is finding a good lead.

That's something that is strongly influenced by luck. I can point to famous researchers who have made entire careers out of this sort of momentary serendipity, actually.

Yes, that is exactly my point. You cannot iterate your way to a serendipitous discovery.

Actually, that's the only way you get to a serendipitous discovery: you make lots of shots, and some of them go in the goal.

> you make lots of shots, and some of them go in the goal

That's not iteration, that's banging your head against a wall and praying for a miracle. Iteration implies methodically and continuously moving towards an optimal solution, not heading back to the RNG after each failed attempt.

We definitely didn't invent systems biology for drug discovery. But Google didn't invent search engines either, I think that we have a fresh and novel approach which is really differentiated by the details.

Another thing I think is different about the industries is that while there are tons of food delivery and photo-sharing startups, there's no really effective/disease modifying treatments for common diseases such as Alzheimer's disease, frontotemporal dementia, ALS, etc, so I think there's plenty of space for any number of new approaches to be tested in the field. I feel it's more worthwhile than pumping billions more into failed antibodies against beta amyloid, as seems to be the trend nowadays.

> We definitely didn't invent systems biology for drug discovery. But Google didn't invent search engines either

This is a flawed analogy, because Google's success is largely attributable to business model innovation (AdSense), rather than any core technology, such as PageRank. After all, Robin Li published Rankdex two years before Brin/Page published PageRank, and we all know how trivial it is to engineer around software patents.

> Another thing I think is different about the industries is that while there are tons of food delivery and photo-sharing startups, there's no really effective/disease modifying treatments for common diseases such as Alzheimer's disease, frontotemporal dementia, ALS, etc, so I think there's plenty of space for any number of new approaches to be tested in the field.

That's because the bottleneck in tech is (a) identifying problems actually worth solving and (b) a business model that will lead to a sustainable company. Achieving success means creating a business and iterating to a position of dominance. There was never any question regarding the technical feasibility of Airbnb's site or Uber's app. It was (and still is) all about things like whether Airbnb could create trust between hosts and guests, and whether Uber could navigate the regulatory landscape.

The biomedical space could not be more different. Everyone with a half a clue knows that neurodegenerative diseases are a problem. Nobody who knows the industry is questioning that tons of customers for your (potential) product exist, or that they'll pay boatloads of money for a treatment or cure. The gigantic elephant in the room is whether your product works. Nothing else really matters.

Biotech startups with a computational method of target discovery that is "different from the norm at Big Pharma" have been around at least since the late 90s. The problem is targets are cheap. Validating those targets is expensive and is the rate limiting step. I have no problems with these companies existing. They should and can tackle some hard problems and develop algorithms and methods that are difficult to pursue in Big Pharma. The problem is that too many are presented as silver bullets that will magically cure xyz, and that's just false.

Yeah, I know what you're saying here, and I don't disagree. I only disagree with the comment that they aren't "doing something new". It's sufficient to do some old thing better in some way that matters to existing customers -- and that's an implementation detail that will never come out of a TechCrunch press release.

We have been actively validating targets, and in fact Alice's PhD work was centered on performing the in vitro and in vivo validation experiments for computational drug leads. We aren't asking for people to believe in magic, and ultimately we hope that our preclinical experiments will reveal whether or not we have something.

It's been the case in nearly every industry that the "small and nimble upstart" rarely wins. Until they do. Biotech the seems ripe for the same sort of nimble innovation we've seen in software and hardware over the last couple decades. Wrote a bit about this here: http://www.sethbannon.com/biotech-in-the-garage

Your article employs the same sort of breathless rhetoric as TechCrunch:

> A smart software developer can build and launch a web or mobile app and get paying customers for under $2,000. The same will soon be true for biotechnology.

Moore's law is a unique phenomenon, and it most certainly does not apply in biotech.

It's true that Moore's law doesn't apply everywhere. In fact, some costs (like the costs of sequencing a human genome) are falling much faster than Moore's law would predict: https://upload.wikimedia.org/wikipedia/commons/b/b7/DNA_Sequ...

1. That graph completely omits the quality of the sequence data being generated (hint: it's pretty terrible).

2. Hyping up cheaper sequencing the way you are is nothing more than what Sydney Brenner referred to as "low-input, high-throughput, no-output biology" (video source: http://thesciencenetwork.org/programs/reading-the-human-geno...). Also see: http://wavefunction.fieldofscience.com/2015/03/cancer-genomi...

3. Here's a more useful Moore's law derivative that's seen in the life sciences: http://www.nature.com/nrd/journal/v11/n3/fig_tab/nrd3681_F1....

Wow. That last chart is really damning. It goes a long way to explaining why biotech has been such a moribund area of investment.

You can sequence as fast and as cheaply as you like, but the big-O on biological research is not our ability to generate DNA sequences.

While cheaper sequencing hasn't really produced much in the way of new treatments (yet), I disagree that it hasn't been productive or that sequence data is of low accuracy. For example, it is striking to see how relatively little was known about Mendelian disease genes before 1990 (see https://www.genome.gov/Multimedia/Slides/GSPFuture2014/04_Mc..., slide 6, as an example for retinal diseases). I think one of the most frustrating things now is that the price of sequencing isn't continuing to decrease.

You bring up another good point about "Eroom's Law" in which pharma productivity is actually decreasing, and that article was actually one of the things that got us interested in trying to do something new. What we are trying to do is to help reverse that trend by doing something different than what large pharma companies are already working on - if we keep doing the same thing despite ever-decreasing productivity, then I don't see how we can expect the trend to change.

I understand that journalists often create a simplified straw-man to contrast to a new and innovative upstart, however as someone who once served as an industry analyst at the intersection of life science discovery and technical computing this article's characterization of how big pharma approach drug discovery seems almost cartoon like in its stereotyping of the discovery and development process at big pharma.

Can you point us to a link that has a more substantive discussion of the approach that you are taking?

Hi there - thanks for the interest! For a more technical dissection of some of the genomics approaches we employ to identify neurodegenerative disease targets, check out these reviews:

http://www.nature.com/nature/journal/v461/n7266/full/nature0... http://www.nature.com/nrg/journal/v16/n8/pdf/nrg3934.pdf

I don't currently have access to those publications and not willing to pay just to read these articles but from the titles, abstracts and citations it sounds like basically you are taking a systems biology and GWAS approach?

The Institute for Systems Biology has been around since 2000. I remember interviewing people in an internal systems biology group at one of the largest pharma companies in the world in 2004 and writing about their work. Admittedly, they were a small unit intended to drive innovation within discovery at that pharma. By 2007 it had been at least tried in most major pharma companies and John Russell was writing about the Awkward Adolescence of Systems Biology.

http://www.bio-itworld.com/issues/2007/sept/cover-story/

Even if your characterization that large pharma prefer a different approach is mostly accurate it is a substantial disservice to the intelligence and hard work of hundreds of scientists who have worked on applying systems biology to the study of disease and treatments.

Maybe I am over-reacting to this article and your video ...but from 2000 to 2004 I heard so many biotech and related software executives tell me that their company was going to solve these same discovery and development problems. Most faded into oblivion, others found a useful niche and a few got targets identified and development started and were acquired by a large pharma to bring them to market.

I sincerely hope that you have discovered a new approach to system biology that will in fact provide breakthroughs. If you have something great let it stand on its own merits; there is no need to misrepresent the hard work and investments of other scientists or the companies they work for.

Keep in mind that this article was not written by the founders of Verge Genomics, but by a journalist who "writes about internet culture, social networks, and consumer-facing technology." So they are probably going to use speech that will make the more science/biotech literate person bristle.

Does YC provide advice to tech companies about communicating complex ideas to not so technical writers and a non technical audience?

YC doesn't provide advice about this in a systematic way, but we have been trying to learn how to engage the lay public mostly from talking to other YC founders. It's hard to communicate the exact details that we want to share, but I hope that the spirit of our ideas comes across clearly.

That was my first thought - my other comment put the initial onus on the journalist for creating a strawman opponent. However, when writing my second comment with a little history of systems biology - including within large pharma - I thought I should go watch the video. The founders basically say the same kind of stuff in their video. So we cannot blame it all on the journalist.

Ahh, I should have watched the video before I had the knee-jerk reaction to be so charitable. They could have just done what grad students are basically trained to do: use your narrow expertise to bullshit people and tie in some buzzwords that grant funding agencies like to hear. I hope I am just being cynical and that YC knows what they are doing here.

Our video definitely was geared toward a lay audience, but we tried not to use buzzwords and bullshit and to explain the science in a more accessible way. I'm curious to know what what bothered you about the video and we will try to communicate better - at least for me, conveying scientific ideas in an understandable way has been pretty challenging and I had hoped we did a good job in the video.

I really appreciate your feedback and of course we have studied a lot of what worked (and mostly, didn't work) in systems biology over the past decade, but the field has shifted dramatically since 2004. A lot more data is available, and experiments are designed much more carefully nowadays. In fact, the thing we worry about most is that the field still isn't rigorous enough and the quality of the data still isn't good enough to translate into druggable targets. And we don't intend to belittle the work of scientists who really laid the foundation for what we are trying to do today, but surprisingly few of the big pharma companies we talked to have active systems biology programs for neurological diseases (we have also learned about others, like at Janssen, that are just getting started). I think it's not a question of if systems biology will lead to new therapies, but rather when the methods and field are mature enough.

Also regarding articles, please email me at jason@vergegenomics.com and I will send you PDF versions to read.

Thanks I will shoot you an email. Your response in this thread is the kind of more nuanced and accurate positioning of what you are doing and why systems biology might now be ripe to make a real difference in the drug discovery phase. It really isn't that hard of a message to convey.

I have been doing other things since 2004 and not following pharma and biotech closely. It appears that what has happened is large pharma invested in systems biology in the early 2000's and by the end of the decade had let it fade. It may also be true that it never made it into specific disease groups, like neurological.

Systems Biology feels a little like AI in the software field. It feels so intuitive that it will one day make a big difference, but there have been many big investments and disappointments over the years, but each time it makes a resurgence we get a lot of advancement in narrow domains. And, it still feels like one day, maybe soon we will get breakthrough's in more general AI and maybe you will lead the way with SB.

A certain confidence, even brashness of youth is ok, but please understand how "tinny" it sounds to people who have been around for one or two boom/bust cycles. Forgive us our cynicism.

> I understand that journalists often create a simplified straw-man to contrast to a new and innovative upstart, however as someone who once served as an industry analyst at the intersection of life science discovery and technical computing this article's characterization of how big pharma approach drug discovery seems almost cartoon like in its stereotyping of the discovery and development process at big pharma.

Journalists often don't create the simplified strawmen, the PR staff writing the press releases for the firm that they are covering do that for them.

Genomic vapor ware.

When will people understand the problem with genomics isn't complexity, it is a lack of raw data.

We simply have not collected enough annotation on what every gene does and how it interacts. Without that information, you have diddly-squat.

"Garbage in, garbage out", or, "give me enough degrees of freedom and I can fit an elephant". It's a reasonable retort given what's on display here.

On the other hand, if you do collect enough data in a particular domain to be able to call some interesting targets, it's straight-forward enough to validate the method. To presume generality (as the scientists probably do not, though as the article and every writeup I've seen here do) is surely foolhardy.

As a side note, if the company wishes to pretty much sell to 'big pharma', why then do they publish these touchy-feely tech-crunch articles at all? Their website is designed for all those web-2.0 ideals of retaining customers and click-through and responsiveness, etc. But I doubt Roche cares about any of those things...

The devil is in the details. Look at all the drug trials that work in mice or even monkeys but not humans.

This isn't 100% wrong, but it's definitely close. To out it vaguely, there is enormous complexity and structure that we can't even model yet.

it might be fairer to say that before we even get to the complexity problem we have to solve the raw data problem.

> But Verge Genomics is able to cure brain diseases 1000x more quickly than big pharma by using algorithms

Guffaw. Tech journalism at its finest. The company sounds great, but I'm not sure Techcrunch is really doing them a lot of favors by just making stuff up.

Right, if they'd said "Verge hopes to be able to cure brain diseases 1000x more quickly" we might cut them some slack. Not to mention that many of these diseases haven't been cured by anyone before, 'big pharma' included, so I don't know how you'd estimate the improvement in time to market...

Considering that the best ALS drugs cost about $1000/month and have results that statistically are barely distinguishable from placebos, there's a long way to go.

My biggest fear is that the folks at YC actually believe claims like that.

The 1000X number should only be attached to "cheaper", and comes from in vivo and in vitro validation experiments we have done; our drug hit rate was almost 1000X higher than high-throughput screens from the literature, and our total spend to get to our one drug lead was almost that many times cheaper than what we learned that pharma spends. We think we are faster as well (our first proof-of-concept study took about 9 months, which is faster but definitely not 1000X as fast), but I think the most important thing is to find something that actually works to treat patients. That's what we want to prove out with backing from YC.

So you've got one drug lead, and haven't gotten anything to market yet that I can tell- are you leads in Phase I, II, or III trials now?

It is highly misleading to use the term cure in this case. You simply have no data to support that your costs are any lower than pharma until you actually show people being cured, and I think Phase II or III is the actual point where you can claim initial cure results, and it's not until the drug has been widely deployed for years without side effects that you can claim it's a treatment or a cure.

It's always great to see new approaches to pharmaceutical development. It's a field that's desperately in need of a better approach.

I do have to wonder, however, when having a focus on neurodegenerative diseases that are largely seen as proteopathies, how useful a method that is presumably targeting only gene expression can be, which is what I assume you're targeting. If the view of diseases like PD as prion-like diseases with perhaps randomly misfolded proteins that recruit others and propagate through the brain is correct, for example, will targeting gene expression be able to do that much beyond treating symptoms? And if targeting genes that are not necessarily

I'm divided on the usual search-of-preapproved-drug approach too. On the one hand, it vastly lowers costs, speeds up trials, and can actually make development somewhat feasible for startups, but on the other, it does really feel like searching through a bunch of drug candidates that have no reason why they'd be effective for your targets. It's unfortunate we don't have a better approach here. And of course, as I'm sure you know, success in mouse models often translates to nothing at all in humans.

I'd be curious to know what sort of genomic data you're actually using for your analysis.

Again, though, it's great to see a startup working in this area, and I'll look forward to seeing what you come up with.

I am a big believer in the prion-like spread hypothesis which seems to be a common thread for neurodegenerative disease, and led a study (in press) that supports this idea in Alzheimer's disease and frontotemporal dementia. Once there is a critical mass, I think it would be challenging to slow the progression of the disease, which may speak to the importance of early treatment. However, given the proven genetic risk factors for these diseases (Mendelian forms, GWAS hits) - which are related to things like UPR, microglia activation, etc - I think that there is a possibility to slow the progression of disease. We also feel the same way you do about the repositioning angle, but it is a nice way to try to get something safe to patients that can potentially be proven quickly and cheaply in a small Phase 2 trial.

In terms of the expression data for diseases, we are doing a systemic review and re-analysis of publically available raw micoarray data from GEO, as well as RNA-sequencing and microarray data that we will publish soon and make publically available. That part is what I mentioned in the video; rather than just looking for differentially expressed genes (which is what most of the associated studies have done), we use network methods we developed to identify what we think are the key drivers of disease. Also, by combining data we get more power to find signal. To match these with drugs, we use a combination of proprietary gene expression data as well as data from large public projects.

Thanks for your comments and wishes!

Hi all! I'm one of the founders of Verge Genomics. We want to change the way drug discovery is done and are passionate about getting much-needed medicines into patients hands within their lifetimes. Happy to answer any questions!

I have a hard time believing that large pharma research labs are limiting themselves to single loci for a given disease phenotype. But if they are, it makes sense why YC would see value in this.

They aren't

I'm going to presume that I don't really understand the company then, because the founders and advisers should certainly know more than I do about drug discovery.

> the founders and advisers should certainly know more than I do about drug discovery

Unfortunately, the operative word here appears to be "should."

We are definitely not experts in regards to pharma (neither Alice or I have experience working in pharma) but our advisors include pharma consultants and senior scientists. We're constantly engaging them and meeting with pharma execs and scientists almost daily to learn what has and hasn't worked, and have received a lot of feedback about how to bring potential new drugs to patients. There's a line between being naive and being jaded and we are trying to stick to being as realistic as possible, and value any new information or feedback, if you know anything that's contrary to our experience.

They haven't for a while. Some of the claims, at least in this story, are way over the top.

Reply on news.ycombinator.com