Intro: SETI@Home

Unless you've been hiding in a cave behind a firewall for the last few years - or, God forbid it, haven't had internet access - then you've almost certainly heard of SETI@home¹. It's a vastly popular distributed computing project which millions of computer users have helped run. Basically, distributed computing means running a program on more than one computer simultaneously to reach the target more quickly. SETI@home took this idea to a new scale, using the popularity of the internet, combined with people's² fascination with the possibility of extraterrestrial life, to encourage more and more computer owners to take part in processing radio data received from outer space by radio telescopes. To do this, United Devices designed a non-intrusive 'Agent' program to run in the background or as a screen saver, so that people could literally take part without any effort on their part.

This, as I may have mentioned, was immensely popular in certain circles. Meanwhile, sceptics like myself pointed out the almost certain failure of the project due to the devastatingly huge odds against receiving a radio signal from an extraterrestrial civilization³. So I have never bothered running the SETI@Home Agent.

A medical alternative

Note: I am not qualified or particularly knowledgeable in medical matters, so if the following doesn't make much sense, feel free to complain. I think I've got the general gist of it across though!

But now, United Devices have started a new distributed computing project which I cannot argue against. The Intel-United Devices Cancer Research Project uses volunteer's computer power to process molecular data, as part of research being conducted by the University of Oxford and the National Foundation for Cancer Research.

Nowadays one in four people suffer from cancer at some time in their lives. The high cost and lack of success of current treatments is a constant thorn in the side of medical research. The last century has seen many astonishing medical breakthroughs, yet the holy grail of cures, the cure to cancer, remains elusive.

UD Cancer Research - The medical bit

Cancer, put simply, is a bug in a cell's programming. Errors and variations are always happening at a genetic level in our bodies; this is evolution at its most basic level. Normally these variations are not a problem, they affect only one cell among billions. But when the variation affects a certain instruction in the genetic code of the cell, the problems start. This particular instruction tells the cell to stop multiplying. During the everyday growth and repair of your body, individual cells will reproduce themselves by division. Once growth or repairs are finished, the cells automatically stop dividing. The instruction to do so is triggered by certain proteins. If the instruction is not carried out due to the aforementioned error, the cell will carry on dividing and growing indefinitely; as far as it's concerned, it hasn't been told to stop. This produces a cancerous growth.

There are many theories about how to stop this growth. The simplest is to cut the cancerous growth out of the body; this can work, but it is difficult to remove all the affected cells - any left over will carry on reproducing.

A more cunning method is to use the body's own 'programming' to stunt or stop the growth. The cells may not respond to the 'stop growing command' protein, but if other proteins are activated they may produce some effect that affects the cells in a different way. For example, a protein named Vascular Endothelial Growth Factor (VEGF) assists in the growth of blood vessels. As the uncontrolled growth of cancer cells requires a lot of blood flow, restraining the flow to the cancer area by limiting blood vessel growth may well slow down the growth of the cells.

UD Cancer Research - The computing bit

Four such proteins are currently being investigated by the project. The proteins are activated and commanded by the introduction of certain molecules, which bind to the protein. However, scientists don't know exactly what molecules do what, or even which are capable of interacting with proteins; and as there are billions upon billions of molecules which potentially behave in this way it is a daunting task finding which the important ones are.

This is where the computing power comes in. Once somebody runs the UD Cancer Research client 'Agent' on their computer, they are sent a batch of virtual molecules and a protein, and the computer runs through all the possible configurations of each molecule, finding out whether one of them will 'fit the lock' and potentially interact with the protein.

Although an average home computer can try out thousands of these variations every second, only a very small number produce results. Once a 'hit' is registered, it is sent back to the server with the next update, for further investigation.

Put simply...

...you can use your computer to help find a cure for cancer.

There is no need to understand what the project is doing; and no need to donate money or effort. All the Agent does is use computing power which you are not using anyway. While you are having a cup of tea or reading a web page, your computer is normally doing very little. Most of the processing power in your computer is only used sporadically, when you are loading a program or performing a particularly complex calculation. So why not make use of it?

Peregrin

05.04.01. Front Page

Back Issue Page

---

¹ Search for ExtraTerrestrial Intelligence
² Read 'geek's'
³ My argument: If the civilization was, for example, 100 million light years away, then they would have had to broadcast a signal exactly 100 million years ago; considering we've only been broadcasting radio signals for less than a hundred years, it would have to be a pretty big coincidence for them to be at exactly the right stage of evolution at exactly the right time and place. I'm not saying that there isn't life out there, just that we haven't got a moth's chance in a light bulb shop of finding it. Disagree if you like but I think we're wasting our time.