finding the expert
|Opened on:||Tuesday, May 3, 2011 - 08:19|
|Last modified:||Sunday, May 8, 2011 - 07:28|
It occured to me that only a very few expert players are needed to solve most of the unsolved protein structures. The key to finding the experts is to have foldit do what 90% of us can do.
An expert should not need to see the protein. Or manipulate it effectively in space. Or perform a rebuild. Or perform a sidechain snap.
The expert only needs to know the goal, and then be provided with a few examples.
foldit needs to be linked to a neural network. foldit needs to become far more intelligent and scalable, to lower the bar for the entry-level player. The tools must become incredibly more sophisticated and powerful.
The more playable the game becomes, the harder it will become to excel at it ( 90% of all players will fall into the largest percentile )
Only an expert will be able to quickly and efficiently evaluate a hydropathic pathway to determine the conformational probability of a polypeptide tending in a certain direction.
The single handful of individuals whom excel at this will be your expert players.