Distributed Computing Applied to Medical Research

Troodon writes " BBCnews SCI/TECH has an item: Screensavers could save lives , anouncing the team up between Parabon Computation and the US National Cancer Institute to apply the idle time of home computers in a Seti@homeesque manner, simulating the responce of cancer cells to potential drugs. The sweetner being the _option_ to receive a payment for your troubles. Other new start ups are jumping on the bandwagon, e.g. Popular Power's choice of contributing to research on Influenza Vaccination, or making a little money with big business. But with these companies potentially looking forward to a healthy cut in the profits of any new drugs developed etc., should us plebs look towards more honourable projects, such as trying to help research into the global warming, that all these boxes dug out retirement are going to contribute to?"

sell your cycles

How much longer will it until companies will sell their IT department's extra cycles on a block?

Seems like a large company like IBM or Microsoft could sell theirs (theoretically they could enforce the company screen-saver/distributed computer on every machine) with a guarantee of X amount of computing for Y amount of dollars. IBM should be familar with that auditing from the mainframe heydays.

Vindictive Attitude.

But with these companies potentially looking forward to a healthy cut in the profits of any new drugs developed etc., should us plebs look towards more honourable projects,

This is the side of the Free Software movement that makes me sick. Theres a vindictive attitude towards capitalism, and especially any capitalism that involves intellectual property.

If some drug company makes money *and* cures cancer using my CPU cycles, I'll be happy. Who knows, it might be my cancer they cure.

seti@home ISP

It could be feasible that a piece of distributed computing software could be bundled with a browser and a dial-up account in a similiar fashion to AltaVista's free service. The big difference would be that instead of banner ads taking up your screen real estate, you have to install a screen saver that crunches data.
The business model would then be:
Company (or research institute) needs data crunched, but does not want to pay for computers ->
Company gives data and algorithms to @home service and pays a hefty sum ->
@home service divies out the algorithms to people across the land next time they login to their free dialup account ->
Data is crunched and sent back to the @home service at a reasonable rate ->
Everybody wins :)

Now somebody is going to get rich off of my idea... it's really painful being this dumb.


"Blue Elf has destroyed the food!"

No.

Personally, even though these guys would pay me and SETI will just suck my bandwidth, I seriously doubt you'll find one of my boxes running any of these sims. (And I'm a med student, so I've got a vested interest in finding lots of new medical miracles.)

Why, you may ask, would I want to waste time looking for nonexistent little green men instead of helping the dying boy who had cancer but is now smothering under a heap of greeting cards?

Easy. We all know exactly what's going to happen when one of the simulations shows something interesting. It'll be snapped up and patented as soon as the data block hits their servers. You? You'll get a micropayment. If you're lucky, they'll mention your name. If you're the first one ever to find something useful, you'll get publicity shots. Royalties from the patent? Yeah, right. You run these screen savers and all you're doing is helping a greedy bastard get rich. (Yes, I'm sure there are some ethical and altruistic biomedical profiteers out there, but my observation is that they're mostly just bastards.)

The IP rush in biomedicine right now is scaring me. The prevailing regulatory attitude seems to be that life in general is just another resource to be locked up and exploited. I don't care how much they're paying per FLOP, I refuse to speed that process along.

Popular Power supports Linux

I'm the CTO of Popular Power [popularpower.com]. Good discussion here, thanks! I really think this technology is neat; we can turn the Internet into a very powerful resource and then use that resource to solve important problems. Our current influenza work is contributing to research that could save millions of lives.

One poster here wondered how good this kind of distributed computer would be at biotech apps. Depends a lot on the algorithm, but for things that trivially parallelize (like random search algorithms, Monte Carlo simulations) it's a perfect match.

Popular Power has been up and running since April. We've had a Linux client [popularpower.com] out for a couple of months now; download it and try it out!

These guys need to obfuscate

Jeez, for something that is supposed to be a distributed system, it sure isn't secure. The data being processed (from popularpower.com) or at least the code for it is stored in a jar file in either a /temp folder or a /tmp folder (depending on win32 or linux install). They are all java .class files in several packages, NONE of which is obfuscated in the least.

Some of the class names:
immune.AntiBody
immune.BCell
immune.Simulator

many others

they all reversed nicely.

Some of the code is pretty cool, I just wouldn't expect very accurate responses when the kiddies get ahold of it.

Global Warming Solution?

"should us plebs look towards more honourable projects, such as trying to help research into the global warming"

Sure, run all your computers 24/7 at 100% use to help solve Global Warming. Makes sense to me.

Fraud...

This is a great idea, but how secure is it going to be? I remember hearing about empty packets being sent back to both distributed.net and seti@home. If people are being paid for this, there is all the more reason to be worried about this - look at the fraud with AllAdvantage (don't use linux and doubt it's avalible for the platform 'cept through WINE - for those out of the know, many people have anti-idlers so they keep on racking up the cash overnight and not looking at the banners).

And this isn't just a relatively meaningless deal like distributed.net/seti - honestly, when compared to people's lives, how much does that matter?

Anyone have any comments?

Similar Product.

Last year, I group of us at school (U of Michigan) did a project to do generic distributed computing... We didn't do anything with the project after the semester but maybe it could be revived now... It's basically a client-server system to apply distributed computing to different tasks..... the URL for the project homepage is here [umich.edu]

$$$

Here's how it works:

1.)2 hours of cpu time : 4 cents.
2.)1 block of data : 25 cents.
3.)1 Cured Disease: 100,000,000 Dollars.

Ok, Count me in.
-Swift ::

Other parallel projects

Is it just me, or does it seem like complex biochemical simulations might not be quite so efficient for this kind of loose, massively-parallel processing? With, say, the SETI data, is makes sense, since there are relatively few operations to be performed on a given data set. To model the action of drugs on cancer cells or the influenza virus, though, you'd either need complete simulations running on each node, or a lot of interaction between them.

I'm not a biochemist, so I can't say for sure, but it seems to me like running a complete sim of that type in a screensaver is a bit, shall we say, ambitious? That leaves parting them out, which is exactly the kind of process that clusters are poor at. If every atom in the simulation is bouncing off of an atom that 'lives' on another user's machine, modem-speed latencies will kill the whole thing.

Priorities

In a recent discussion in this space about SETI@home, one of the threads was something along the lines of:

How can we spend so much time and so many resources on this when we have so many problems here on Earth?

To which an astute poster replied something like:

What resources? Extra computing cycles? When they find a way to feed people or cure cancer with extra computing cycles, that will be a valid argument.

Uh-oh.

So do we now all have a moral obligation to dump SETI@home?

-

Re:Vindictive Attitude.

This is the subject of much debate in the press, and unfortunately there is way too much politics for anybody to really be informed either way. Just when you think you've found an objective study, there is usually a money trail leading someplace, and the leftists are no less guilty of this than the rightists.

Your argument is similar to the left wing argument. The counterargument is that without profit motive the drug won't be developed to begin with.

The counter-counter argument to that is that we are placing economics above public health. The counter-counter-counter argument to that is that economics *is* public health; i.e. a world where researchers at Pfizer have to live paycheck-to-paycheck is actually less healthier (at least for those researchers).

In the final analysis, we need some way to determine the proper balance, or "equilibrium" between the profit motive of the drug company and the profit motive of the drug company customers.

Under an the extreme left approach, socialization results in a decreased ability for the company to deliver a drug. Under an extreme right approach, economic circumstances prohibit the lower classes from receiving the drug.

Now, the trick is to determine the situation that maximizes delivery of the drug, which lies somewhere in the middle of these two extremes.

There is nothing particularly honourable about giving drugs away if you can only do it for a little while, just so you can go bankrupt and/or cut research staff. It is far more honourable to maximize delivery. So far, the limited life drug patent is the best method we have come up with to do this.

Re:seti@home ISP

Better yet, they offer DSL, and throttle your bandwidth based on the amount of data you process per month. Build a home cluster, run their software, and they'll connect you at T-1 speeds for free... Anybody wanna start looking for some VC?

Cool business plan, but...

The idea is cool obviously, but frought with ethical dilemmas.

Does the company owe the person whose computer discovered the drug anything? If so, someone is eventually going to win a computational lottery. If not (at least in U.S. versions of this) you could easily get some eldery person unable to afford the very drug their computer researched.

I can easily see the day when picking your screen saver will be almost as challenging as picking the charities you choose to support.

Re:No.

We all know exactly what's going to happen when one of the simulations shows something interesting. It'll be snapped up and patented as soon as the data block hits their servers. You? You'll get a micropayment. If you're lucky, they'll mention your name. If you're the first one ever to find something useful, you'll get publicity shots. Royalties from the patent? Yeah, right. You run these screen savers and all you're doing is helping a greedy bastard get rich.

Yeah, right, like you would actually deserve fame and fortune for the extremely challenging task of running a program. It only seems fair that the creators of the program would get that, no?

(Yes, I'm sure there are some ethical and altruistic biomedical profiteers out there, but my observation is that they're mostly just bastards.)

Tell you what -- next time you get some sort of infection, don't support one of these greedy profiteers by buying antibiotics. No, die a painful and gruesome death, and by doing so, send a strong message to these bastards.