Besides the beautiful canary-like colors of this diagram, a biologist might notice a problem is here. Something is wrong with the shape on the right...
On Friday Dr Amy Keating spoke to us about this.
She said that humans playing games can significantly aid a computer in creating the minimum energy, most stable protein shape.
Humans can outthink computers in path-seeking because they can see the forest, not just the trees, or in this case, the final overall protein shape, not a local shape. The naivete of a protein folding computer, or any computer which seeks a path from Point A to Point B, is that it considers only paths that seem to lead from where it is to the next local desirable point. It might travel to ever lower, stabler fold shapes but then stop at what it thinks is the best, lowest point when really it just looks like the lowest point locally. The computer is shortsighted, thinking of only immediate goals.
Protein folding takes a human brain to sense whether or not the d Its a short-sightedness on par with some national political institutioiscovered shape is really the lowest point or if it is just a local minimum.
Keating's analogy for the trapped computer program was to think about a computer trying to guide a robot down a hill. It is easy to imagine the robot seeking ever-lower points, only to come to rest in a crevice that is still halfway up the mountain. I know this ski-resort analogy stuck with the students because they were still using it the next day.
Though I still tend to dislike algebra, I do actually find the constancy of mathmatics and patterns I encounter as oddly comforting- science of the human body doesn't always go according to logic but math always does. My way of thinking causes me to see God in the sheer complexity of the human body and yes, even in the greatly loathed algebra.
Posted by: Denise Spring | January 15, 2010 at 12:48 PM