http://www.nytimes.com/2016/02/23/health/a-do-it-yourself-re...
Featuring the Open Insulin project. Many of the supporters of that project came from HN.
I wish them luck, but c'mon. Biopharma is not IT where a smart kid can compete from his garage. It just isn't.
The NYTimes article says "the hackers hope to be able to demonstrate the technological feasibility" rather than manufacture a drug. But everyone knows that it's technically feasible -- the challenge (and cost) is in the doing.
Well, I am currently producing one early-stage (xenograft) drug candidate literally in my garage, and have made two new molecules based on the structure, also in my garage, about to be submitted to structural characterization (and eventually, in vitro studies and maybe even xenograft)... But yeah, I do have a PhD in chemistry, and spend 10-20h/wk on the project.
(also with the caveat that all my studies hit a "brick wall" of can't do in the garage when I have to start worrying about GMP).
>Extensive pharmacological testing in cultured cells and animals is much, much harder.
And that's why I'm contracting CROs for those parts. I guess technically that's "taking out of the garage", but I never said the whole thing is being done in the garage!
True, that's why I said "at our company". It differs everywhere, as you mention. My only point was that we screen thousands of compounds, and it's only after a compound has shown desirable activity, safety, ADME, PKPD, etc, that it becomes a real "candidate". Not everyone realizes this, and thinks you can synthesize a compound and the next step is to file an IND.
Your point about CROs is a good one, and I thought of it after my post. We use lots of CROs, too. And there are obviously lots of successful virtual companies that don't do any wet-work at all. But CROs cost lots of money. Not really cheaper than doing it yourself, just more convenient sometimes. Virtual companies can work because they have millions in VC.
Anyway good luck with your endeavor. I don't mean to sound cynical or negative, but not everyone in IT realizes that the two fields are qualitatively different in many ways and I think the differences are worth pointing out since they're often overlooked around here...
OTOH, I'm getting xenograft for just about $30k, which I think is about right.
> not everyone in IT realizes that the two fields are qualitatively different in many ways and I think the differences are worth pointing out since they're often overlooked around here...
Totally agree.
It isn't today, but it may be at some point. We tend to underestimate the effects of monumental shifts like we've had in IT. This project represents the first steps in that direction.
> The sort of thing that a single undergrad could do in a few days.
Even if this were true, the OpenInsulin project doesn't have a large institution's lab facility. Nor does the project have a government grant as benefactor. They are operating only off crowdfunding. It's obviously not enough to get through the FDA approval process, but IS enough to construct a roadmap to that end and make headway on a functional insulin producing organism. The project already has E. coli making proinsulin. The experimental focus has now moved to methods of purification. They are also investigating the feasibility/cost of getting this through the FDA approval process for biosimilar molecules.
I think the real magic of this project is its open nature, which I find to be lacking in the scientific community. This is research that anyone can make meaningful contributions to. Science (specifically biology) can learn a lot from open source software.
But what makes pharma expensive and time-consuming is the regulatory process. Producing drugs is easy - meth dealers do it all the time. But spending a decade getting your drug through the federal system is expensive, and there's no technological shortcut. I'll take more interest when they stop "investigating the feasibility/cost of getting this through the FDA" and publish a plan to actually do it.
I just don't understand what this means here. Cloning a gene and expressing it in e.coli is something that has been routine for decades. Literally. These simple experiments won't provide a roadmap to do anything that we haven't known how to do for a long time. It's like a crowdsourcing project to raise money to write a "hello world" program. Seriously.
Open-ness is great. But "meaningful contributions" are generally made at the cutting edge of scientific research. The experiments described are nipping at the heels of science that was worked-out in the early 80s, so I fail to see how a meaningful contribution will be made here. Open source is great, but making headway in drug discovery or development is a Hard Problem and this doesn't strike me as a serious effort. Just my two cents.
Yeah, it's been done before but with different variations. To do so in a way that isn't locked up in IP is worthwhile. And it's a challenge that faces an uphill battle in terms of incentive and payoff. The crowdfunding covers that risk capital gap that would otherwise make this impossible. You could say their approach is a road not taken.
I was also struck by the money involved. ~$15k will pay for an experiment or two, as they indicate on the website. It takes many thousands of experiments to get anywhere in Biopharma.
So... it's a nice start, and maybe a worthwhile challenge, but it hardly seems like a serious effort. And not everyone realizes that, so I thought it was worth pointing out.
It's easy to be a cynic on HN, it's harder to think critically about why someone else would believe that it could work. You don't really get rewarded for empathy on here.