Tag Archives: proteins

Firas Khatib – FoldIt


Add to iTunes | Add to YouTube | Add to Google | RSS Feed

Firas Khatib is a postdoctoral researcher in David Baker’s Laboratory in the Biochemistry Department at the University of Washington. He is currently trying to cure cancer by getting people to play online video games, specifically Foldit, but has yet to break the news to his grandmother who would not approve. Firas received his PhD in Bioinformatics from UC Santa Cruz in 2008 where he gave the graduate commencement speechfor the school of engineering; luckily they still gave him his degree.

The University of Washington are investigating whether the brainpower of humans worldwide can be brought to bear on critical problems posed in computational biology. The long term goal of this project is to utilize the combined power of humans and computers in order to build accurate models of disease-related proteins by introducing a new approach: distributed computing driven by human intuition.

The problem of accurate prediction of protein structures remains one of the most important unsolved dilemmas in biology. Determining the correct structure of a protein helps researchers successfully target them with drugs. Currently, experimental methods for solving protein structures result in accurate, high-resolution solutions, but cannot keep pace with the quantity of information. Developing new computational techniques to produce experimental-quality models is one of the most important problems in computational structural biology.

While progress in high-resolution protein structure prediction is being made, particularly here in David Baker’s research group using our Rosetta prediction algorithm, consistent computerized production of models with atomic level resolution are still out of reach. Even with the current supercomputing and distributed computing power available, the protein structure prediction field is still years away from being able to predict the 3D structure of any protein whose genome has been solved. A novel approach to this problem is the use of human puzzle-solving intuition and pattern recognition skills in conjunction with computers running the Rosetta algorithm.

Foldit is an interactive Rosetta-based program that was recently developed by the Baker Lab and University of Washington’s department of Computer Science and Engineering. This program allows users around the world to directly manipulate protein structure models on their home computers in real-time using Rosetta’s energy function. The ultimate goal of Foldit is to use results from the computer program to improve protein structure prediction algorithms by capitalizing on human 3D problem-solving skills. Human users can recognize that certain moves they make might temporarily result in a worse Rosetta energy score, but lead to a better overall energy score. Humans have the advantage of awareness that often the only way to get to the correct low energy state is to go through many high energy conformations. Foldit users are also able to share their solutions with others around the world, taking advantage of the strengths that different people have.

The more knowledge we have on how proteins fold, the better equipped we will be in the fight against protein-related diseases, such as cancer and HIV. Foldit can contribute to this global fight by giving worldwide participants an immediate and relevant application for scientific knowledge, and hopefully reinforcing motivation to understand the underlying science.

Want to embed this video on your own site, blog, or forum? Use this code or download the video: