Posted: Sep 23, 2011 4:07 am
laklak wrote:"Computer simulations". Lol. How many people here have every actually designed and written a complex software system? I don't mean some web app or DB mining system, I mean a real, live, multiple-linked-mainframe, thousand-man-year system? If you have ever managed (or even participated in) a project of that scale you would never, in 10,000 million years, trust your life to a drug trial conducted in a simulated environment. We can't keep fucking Facebook up and running on a daily basis, let alone model an incredibly complicated biological system that we don't even fully understand. It's utter nonsense, complete bullshit.
There is no system out there and no system on the horizon that will do that. How in the world can we do anything of the sort until we understand, down to a molecular level, exactly what the human body will do under any given set of inputs? FFS we don't even know how most of the medicines we use every day actually work. We have no clue. Just look up common drugs in the Physician's Desk Reference - I'll wager you will be stunned at the number of entries that say something like "the exact mechanism is unknown".
It's a scam, hokum promulgated by people and organizations with a political ax to grind.
The argument also ignores the fact that most drugs and experiments are actually first run through a computer simulation. This is often done as a proof-of-concept trial, but these trials are usually done before any animal testing begins (using the most sophisticated software and computer equipment that is available). The next stage of trials is to test the concept using in-vitro testing; where they isolate specific cells or tissue mass and see what happens when they introduce their drug. At this point we find that our results differ wildly from the computer simulations. The next step is animal testing; and here we often find that our results wildly differ from in-vitro testing and now don't even resemble the results found in computer simulations. Then we go from animals to humans, and whilst there can be significant differences in our results, most of the time we will have successfully used the data to identify ideal dosage requirements and any negative complications that can be avoided.
The idea that we can use computer simulations instead of animal testing is ridiculous though.