Slashdot Mirror
Open Source for Biotechnology
LarsWestergren writes "The Economist claims that Open Source is such a success for software development, the model should be used more often in areas such as biotechnology and bioinformatics. The similarity between open source and the academic process with their 'you share, I share' principles is shown by the human genome project. The paper argues that this process should be used for instance to developing medicines unburdened by patents, useful especially for third world countries or diseases that affect relatively few people, where medical corporations have previously thought that the cost of research have not been worth it."
Open sourcing discoveries in bio-tech would lead to reduced research costs, reduced development times and ultimatly reduced prices of drugs.
..GNU for biotech
It will also by extendtion lead to more competition in the bio-tech industry, which can only be a good thing. And it will lead to more consumer scrutiny of what were popping into our bodies.
This is a good idea all round. Except of course for the biotech monopolies who...
[censor type="DMCA" excuse="Subversive,complaint"]
[/censor]
May the Maths Be with you!
I've been saying this for ages - we need a bitoechnology GPL. In other words, you are free to use the technology and incorporate it into your own research/product developments etc., but if you distribute a product that uses this, your process must be made available under the same licencing conditions. So if I invent a process that's useful in producing a wonder drug, anyone can use it, but all other aspect of this wonder drug must be available for others to improve upon.
43 - For those who require slightly more than the answer to life, the universe and everything.
Open-source software? No problem. Unencumbered research? No problem, and I assume non-profit organizations or government sources are paying. However, as I like my drugs tested before taking them, who other than these non-profits will pay the cost to test the proto-drugs?
No patent protection = no profits. No profit = no investment, and no desire to fund tests.
And then you need to create the drug itself. That takes another many years of experimentation. And then you need years of clinical trials. Then a manufacturer needs to then be found.
And someone would propose that drugs be created using an open source process? What would be the incentive of creating drugs or getting the education to do so? This isn't Linux, it's a complicated process of creating a drug for a human. Get it wrong, and your monitor refresh rate is off? No, people die. This is clearly just a pipe dream.
Hoist Number One and Number Six.
Wikipedia is another example of open-source-like methods being applied to a non-software area. Only time will tell exactly how successful it is, of course.
"How to Do Nothing," kids activities, back in print!
The simple truth is that patents really are a lie about free market economics. They treat it like it's a physical property, but it's not. If millions of people use my car it deprives me use of it in a serious way, but if millions of people use the same invention - then just the opposite happens. The inventor is not only able to keep and use his original invention however he wants, but also now has huge forces contributing to it's improvement.
... they paid for those slaves God blessed, surely that alone shows slavery is good, and the negros have been saved from their barbaric condition" ....
... etc. - it really causes one to think.
If the government gave someone a monopoly on making cars, because they didn't have an incentive to make cars when other people can make them too - most of us would see that as crap. Market share isn't an inherent property right. If the government gave someone a monopoly on growing oranges, on the premise that they wouldn't have an incentive to grow oranges if other people could too - most people would see that as crap too. But for some reason, that logic breaks down when it comes to invention.
Finally, looking back on history to paraphrase "look at the great wealth and prosperity of the plantation system, the grand architecture, the vast and rich land, the free markets
I wish I could say that patents are causing less harm, but when they recently lokcked out 10's of millions of Africans dying of AIDS from getting generics because "they had no incentive", because patents are "a property right", becasue "the wealth of the pharmasutical industry in the US is proof that patents work"
Communism and Open Source do sound similar (in their principles). The parent (was probably a troll) but only because they knew that everyone would see them saying "sounds like something bad to me" instead of "sounds like communism to me." Communism isn't necessarily bad. I was thinking in the shower (before I saw this article) that communism is a set-up where everyone benefits by the greater good of the community. This works by making sure everyone can benefit from the efforts of individuals. Sounds like Open Source doesn't it?
the drug company multi-nationals, I'd like to strongly dissuade everyone from pursuing this idea. If the "people" are free to concentrate on unpatentable, abandoned and unprofitable medicines in some sort of collaborative effort, this will severely hamper our efforts to develop ever faster erectile dysfunction medicines, baldness cures in a pill form, medicines for newly created social "disorders", drugs to strip the carbs out of everything (or proteins or whatever the new black is) and have people pay top dollar for them. Stop rocking the boat and someday we'll find a cure for something (as long as it's profitable).
Then all the technical superiority of the western world is based on communism? Because that technical superiority is to a large part based on research following this principle. I doubt that research would be as far as it is were this principle not followed.
Indeed, one of the main measures of scientific success is the number of publications (unfortunately not taking into account the quality of those), that is, you are considered a better scientist if you added more to the community. Indeed, that measure can be crucial to your career (if you don't publish your results, i.e. don't share, then you'll never get far in the scientific community).
Note that also the incentive of the patent system was to encourage people to share their knowledge. You cannot patent something without publishing it, and that means after your patent expires, it's free knowledge for anyone to use.
The Tao of math: The numbers you can count are not the real numbers.
The benefit is that for people who know next to nothing about coding, or don't want to code, are able to use the software completely free of charge and be able to modify it to suit their needs.
This is the benefit I was pushing for Open Source Biotech.
Buy Steampunk Clothing Online!
I don't think this will work, and let me tell you why. First off, let me preface this by saying that my wife is a soon-to-be pharmacologist, so while I may not have any firsthand knowledge of this, she knows what she is talking about.
1) The cost of research for pharmacology is infinitely (no not literally) more expensive than it is for computer science. In most research for CS you just have to pay for cost of equipment (basic computers typically costing a hell of a lot less than the specialized machines used in development of medicine) and the salary of the researcher. A lot of CS research can be done by one person. For pharmacology you have the cost of equipment (or even the USE of it, sometimes they have to rent time on more uncommon machines; this happens in CS as well, but not nearly as often since it's mainly for the processing power) as well as the cost of the researcher AND his/her assistants. It's almost impossible to do good research in medicine by one's self because of...
2) It takes freaking forever. The number of steps required to find out if a proposed theory for a molecule even has a chance for working is phenomenal. My wife has spent the past few months trying to see if a certain molecule will bond with an AIDS neutralizer. Mind you, this is just the first step. Even if this step does work (which they don't know yet) they don't know if this molecule will a) bond with the aids virus b) will it bond long enough to neutralize? c) if it does bond, will the neutralizing agent be able to reach the virus? or will it be blocked by the bonding molecule? And the list goes on. No pharmacologist who does this for a living is going to volunteer even MORE time out of their lives for no pay. So we'll pay them right?
3) Funding. Right now almost all pharmacology is financed by companies that already have patents or by third party investors. These people invest money into these projects because they expect a profit as return. Yes, I'm sure they also care for the well-being of others, but they do need to recover their costs if a drug succeeds. A vast majority of projects fail, which is why a lot of specialized medicines cost so much. These companies need to stay alive in order to do more research. And don't even talk to me about Federalizing the research. That would be pretty much the dumbest thing ever.
I'm sure there are holes in my argument, but hopefully this will at least provide food for thought and further discussion. Basically, I just don't see it happening.
--------
This isn't the sig you're looking for. Move along.
1) Developing new drug products requires substantial, very expensive facilities, while the hard costs of software development are very low.
2) Drugs must go through a long and expensive testing and regulatory process before being released to the market. Open Source software simply wouldn't exist if it cost millions of dollars and took several years before you could release it.
3) There are massive costs associated with product liability in drugs - no one would give away software if the same liability exposure existed.
4) For every drug that makes it to market, there are dozens to hundreds that don't make it through the process but incur the costs of development anyway. The unsuccessful attempts are subsidized by the successful ones.
While I think that the sharing of information in biotech is generally a good thing, I don't think the economics mesh with a software-like "open source" model.
Well, like everything, this is a complex issue. A lot of data is already "open". You can go to NCBI and download the entire genome of SARS or Bacillius anthracis (Anthrax) if you so wish.
Also, if you are creating bioinformatics tools on Federal funding (NFS, NIH), a lot of times the stipulation is that the source code must be made available. This makes sense because your peers has to make sure that the way you did your calculations are actually correct. If people are to use your data or program in their publications, your program had better be correct. Many times there are no way to tell except to look at your source code.
But before we talk about open source, the real issue is standarization of formats. Bioinformatics is like a jungle right now, and every one has different formats for describing the same thing. NCBI has their formats, the europeans have theirs, and it's a terrible mess. I just spent the past week writing code to parse PDB files. This format has been around for ages, and is so inadequate. File formats designed by biologists do not lend themselves well to compuation. What we need right now is an open standard, based on XML and open APIs for parsing these standard files. This will go a long way towards information sharing, and can save a lot of duplicated effort.
"Glory is fleeting but obscurity is forever" - Napoleon Bonapart.
Buy Steampunk Clothing Online!
Does OpenGL make understanding discrete mathematics any easier?
Biotech is hard. It isn't something you just pick up and do. Making it open source wouldn't make it any more accessible to non-biologists. Similarly, whether a program is open-source or not has virtually no bearing on how your 'average' user uses said program. John Q. Webmonkey isn't going to derive any value from the Apache source code unless he's already a competent programmer.
Open source will help make work easier for biologists, but "the people" won't have a damn bit of use for it...unless, of course, they go through years of study and training first--at which point, they're biologists.
Which is probably why something like this will never be allowed to happen now that people have seen how successful open source is.
Oh, for pete's sake--don't be such a fucking cynic. It's not a sign of some deep wisdom, it's a sign of laziness. You're basically declaring that you're not about to lift a finger in trying to make things better, since you think it'd be a futile effort, anyhow. Here's a clue: humanity has dealt with power-hungry tyrants and money-grubbing shysters since the dawn of civilization, and yet somehow we've managed to progress beyond pointy sticks and thatch huts. You're nutty if you think that the little guys and the altruists have it harder now than they did before.
There are people who make a difference on the world. These people generally do not kvetch about how it's not worth even trying, seeing as The Man will just put 'em down, anyhow.
Obliteracy: Words with explosions
25 years ago one of the better business minds said the same things about the brightest coders in software. Go check out the open letter from Bill to the hobbiest coders. Software costs a ton of money, one wouldn't have expected the brightest minds there to just give a way their research for free. Bill Joy is arguably one of the biggest innovators in the software business, and most of his ideas (SPARC, vi, BSD TCP/IP stack, c shell, aruguably JAVA) have been free since he first thought them up. There are a tremendous number of parallels betweeen the drug industry and the software industry from a financial perspective. Both require signficant R&D to develop new products that then require almost no cost to produce (what chemicals go into drugs and how much do those cost). Also plenty of software carries lives on the line, software controls the ATC system, powerplants, those drug distribution machines, your financial life (imagine if you were to be erased from all your electronic sources of cash-how long could you survive?).
Degaussing scares the bad magnetism out of the monitor and fills it with good karma.
I agree, I'm not Bill Joy preaching the end of the world but there's something to be said about guarding some secrets. All it takes is someone posting their new virus on the internet and 'open' sourcing it for some terrorist somewhere to start developing a really nasty world killer. I think the way it is currently is pretty scary, we've got people basically building new viruses without first understanding how to create a vaccine or drug or other remedy. And there seem to be a lot of people who want to play God out there, it only takes one.
To really understand how a human works takes more than four years of university level study.
Maybe I missed something, but most people off the street would have no clue about how to code a linux kernel, much less keep their computers from being spam servers. And last time I checked, most programmers aren't exactly GED cases. Yes, years of CS training, while not requisite, are certainly the norm among the best code writers.
What would be the incentive of creating drugs or getting the education to do so?
Ahem, sounds almost identical to most closed-source companies' arguments for years? Lo and behold, open source now has M$FT on a run for their money, literally.
This isn't Linux, it's a complicated process of creating a drug for a human.
Being a biological researcher myself, who happens to also be facile with computers (a rarity, believe me), I'll tell you that the difference is small. Trial and error, code and debug, it's all the same. The only difference is the time-frame. Biotech "programmers" just have a bit more leeway for losing days of work on an error than a computer counterpart would. It's the nature of the beast.
Get it wrong, and your monitor refresh rate is off? No, people die.
This argument is stupid. A drug is a tool for combating disease based on technological invention. Just as the EKG monitor in the ER that runs on UNIX is a tool. Not to mention the prehistoric plants that were kind enough to die 50 million years ago and convert into combustible hydrocarbons to power an ambulance. All technology saves lives.
In Soviet Russia, all our base are belong to you!
Yeah, you know absolutely nothing about the biotech world. Biotech drugs can be enormously expensive to produce. Bioreactors to manufacture antibodies cost millions of dollars. Open source has worked because computers are cheap and easy to come by, bioreactors and various other machines (spectrophotometers, ELISA readers, etc.) are ungodly expensive. Suffice to say no lay person can afford to produce biotech drugs, so stop trying to draw a parallel that doesn't exist.
Absolutely not. The R&D costs for software development are almost entirely* for salary. That's why developers with commodity hardware and software at home, willing to code for free, can make significant contributions. Salary costs in biomedical research are trivial compared to other research costs. That's why I'm a biologist, but I write code as a hobby instead of conducting clinical studies at home.
But, as you say, maybe I'm wrong the way Gates was wrong. Anyone willing to step up and prove me wrong?
* Plus, of course, the costs of testing and usability studies. Do you want your chemotherapy to have the level of refinement of the Debian installer?
What I'm listening to now on Pandora...
I had the same problem, but I'm wondering if the confusion is between "open source" and "done for free". I don't think the way the article was written helped much.
For example, they cite the human genome project. This work was done by paid individuals using millions of dollars worth of specialized equipment and going through millions of dollars worth of consumables (tubes, reagents...) in the process. The work was not done for free (we paid for it) but the information gleaned is open for all to see.
In the end, I'm left wondering what exacly the article is calling for that isn't already being done. Other than more of the IP being placed into the public domain.
I wouldn't say all along. America has been capitalist for most of its history, IMHO. But yes, there is definitely a small gradual trend towards socialism these days (ignoring the current buffoons in power). But I think this is a natural trend in all the rich countries. First y'all go thru a phase of hard-core intense capitalism and then when you've finally started reaching a certain level of wealth, the upper crust starts giving back to society, even if it is a trickle. But I think starting out socialist is a BAAAAAD idea for any new nation - i.e. India. It started out socialist and is NOW moving to the fre-market. Try breaking the entrenched attitude that the Government is there to do everything for you, and that you must accept what is given. The one thing that a capitalist economy will give its citizens is a good work ethic - i.e. work hard, and only deal with the govt. when things go wrong. But in socialist countries, people develop a dependent mindset, and this totally screws you over. "If you give a man a fish, he doesn't go hungry today. If you teach a man to fish, he never goes hungry..."
My Favourite Meme
Not really. It is true that the software is widely distributed (and packaged in a handy interpreter!)
But it's rather aggressively copy-on-write; changes generally show up in the child rather than the parent
There's even a government program to try to stamp out self-modifying code!
So: widely distributed, yes. "Open source": not hardly.
To what extent would public universities be willing to foot the bill? After all if the drugs go into the public domain and people get cheap treatments, that is a direct benefit to society. It would seem to be in line with universities' mission. Whether funding boards, etc, etc, can be convinced to spend the money is of course another question entirely, but a few high-profile successes would probably bring them around.
In Soviet America the banks rob you!
That's where they lost me too. Open Source software works because the individuals involved already have all the equipment they need at home (a computer, some software, and a lot of time), and there are really no consumables to account for (unless you count the caffeinated beverages and junk food). Then you get into the cost of the equipment which can range from thousands to millions...
The truth is that much of the research does need to be done in "wet labs" which means real costs Its not like everybody has the equipment necessary to actually synthesize and purify a new drug in their living room, much less test it.
I think the description of the genome project being like an open source project is somewhat misleading. The people involved were paid, and used equipment paid for, by grants from various institutions. Yes the information was pooled for the common good, but its not like these people did the work at home on their free time. Essentially as "open source" as publicly funded research has been for decades.
True, then you acknowledge that it would be AT LEAST 20 years before this could even be considered feasible, and to be honest there are hundreds of times more people involved in pharmaceuticals and biotech now than there were computer people involved back then.
If you want to be involved in any sort of real decision making in drug discovery process, plan on starting with a PhD or MD.
That's fine so long as the analogy is valid, yet I'm not convinced. Tell me why opensource software development and manufacturing are comparable to drug discovery and manufacturing and I'll listen.
Being a biological researcher (whatever that means) does not make one an expert on the pharmaceutical business.
Anyway, I don't buy the analogy. To do drug development you need experts at identifying candidates, experts at drugging candidates, experts at doing the in vitro functional validation, experts to do the in vivo validation and toxicity testing, and all this before the drug even sees a human. These experts are all teams of Ph.D. and they use equipment that you and I can't purchase and store in our basement (unless you are secretly batman). Then there is clinical testing and development of product--figuring out how to deliver a compound internally is its own science--manufacturing and education (marketing).
And sure, some of this is shared with software development, but most isn't. Capital expenditure is enormous. Committment and risk taking are also part of the equation (bear in mind that something like only 1 in 50 drug candidates makes it to market). In short, you have to be willing to fund a research program 10-15 years before you can start to recoup losses.
Maybe if you could convince some of the big pharmaceutical power houses that there was a business model in there somewhere you could interest them, but this isn't something you will ever pull off grassroots style.
So what exactly is the business case for someone like Pfizer or GSK or Merck to start doing things differently???
When used correctly? Maybe. Remember though that "drug" is just a fancy way of saying "poison." You darn well better know what it does in which doseages before you start screwing with people.