HIV design challenge
One of the biggest challenges facing protein design today is to model protein backbones. Unlike prediction, where a sequence is given, a design puzzle has more hidden traps since amino acids can change, allowing multiple (potentially false) answers to nearly identical problems. We are not at a stage where we know definitively how to choose the solutions that will work when produced and tested in the lab. Historically the complexity of a design problem is reduced by holding the backbones fixed. With the FoldIt game, we are asking users to sample flexible backbone designs with score-guided intuition to tackle this problem: how do we build one protein segment at a time under the constraint of a native scaffolding while maintaining the "foldability" of a sequence? Similar to prediction puzzles, we set up scores based on known metrics for assessing the quality of these models, and ultimately try to correlate these measures with experimental data to understand the underlying design principles, while iteratively improving them at the same time.
Using GP120, the HIV protein responsible for entry into host cells, as a model system is significant in that 1) it remains a viable candidate as an AIDS vaccine, and 2) it has an unusual topology. Having the tools to understand and engineer this molecule would contribute greatly to both AIDS research and protein engineering. Among several mechanisms used, the molecule uses a number of variable loops to distract the immune system from forging an effective response. In a nutshell, GP120s are located on the very outside of a viral particle, the envelope, and when the immune system sees it, a wave of antibodies are produced to try to neutralize the pathogen. The problem, however, is that although we make antibodies against this molecule, they are mostly directed to attack loops that are not related to the central machinery (which is hidden) responsible for invading host cells. To focus the response, our strategy in designing the vaccine is to expose the elements directly responsible (marked with CD4bs in the attached figure) by creating viral free proteins that resemble GP120 but lack its cloaking machineries (by editing structural regions marked with A,B,C,D in the figure, for example). In other words, we are interested in trimming away these loops while preserving the area on the surface vulnerable to neutralizing antibodies. We are hopeful because antibodies that can neutralize a wide variety of HIV strains do exist. The idea is to create a "mold" based on the known broadly neutralizing antibodies and present it to the immune system for the production of similar antibodies. The term "reverse vaccinology" has been coined to describe this strategy -- we know some antibodies work; now we try to produce copies of them by guiding the human body to make them.
As described previously, maintaining a protein structure while doing extensive remodeling work remains a challenge. However, one can intuitively imagine "protein sculpting" being useful in many applications of protein design. Besides the AIDS challenge presented above, we can apply the same shaping strategy to improve enzyme actives sites to make them more active, to alter cellular signaling by modulating the strength of proteins interacting with each other, and to create protein chimeras by shaping different parts to fit together for new functions, just to name a few. We are starting to address these problems with FoldIt. Designs that are judged plausible will be systematically studied by actually making them in the lab and testing for their (hopefully improved) functions. The goal is to learn enough about proteins through this process: to fundamentally improve our understanding of protein biochemistry and potentially create a vaccine or an enzyme along the way.( Posted by possu 81 2367 | Thu, 05/21/2009 - 18:20 | 1 comment )
Quest to Native
We have recently noticed that for some puzzles, althought the scores are significantly improved, the native is in a slightly different place then where the game community has been exploring. Currently we would like to find out of the reason why the native protein structure remains elusive in those puzzles. There are two main possibilities:
- the native structure is not reachable with current set of game tools
- we did not provide sufficient intuitiion behind the game mechanic to direct the game exploration in the direction where the native structure is hiding.
To find this out we will be posting a few Quest to Native puzzles, where the native structure will be shown as a ghost. The challenge for the community is to see if you can find the significantly better score that is around the ghost. In the process we would also be interested in whether this process gives more intuition about how native structures can pull a much higher score.
Zoran( Posted by zoran 81 2367 | Fri, 05/08/2009 - 15:23 | 18 comments )
Help Needed: Personal stories of learning through foldit
Over the course of developing foldit, it has become increasingly apparent that the game environment although always imperfect and evolving, is not only a vehicle for scientific discovery, but also for learning about proteins, and biochemistry in general.
In order to fully develop this alternate, but perhaps just as important outcome, we are writing an NIH grant to, over the next 2 years, develop an educational version of foldit to be deployed in high schools, or perhaps even middle schools worldwide. Together with learning scientists who are joining our effort, our plan is to quantitatively and qualitatively evaluate the effect of such a game on both the comprehensive understanding of biochemical underpinnings of life, but also to gauge the effect of a game like foldit on the perception that science is a cool and worth pursuing.
The grant is due this Wednesday (ugh), and we would love strengthen the proposal by including the personal stories from you about how foldit community enabled learning. So please send them in response to this blog, or email them directly to me (email@example.com).
Thank you.( Posted by zoran 81 2367 | Tue, 04/21/2009 - 03:22 | 1 comment )
All-Hands Update and its relation to the Grand Challenges
We have reached the final stages of our study of the effectiveness of human reasoning on the structure prediction problems. This final stage, as it turns out, will also be critical to some of our findings. Let me briefly explain what we have found so far. On well-structured localized problems foldit players do very well. On more complex problems players also do well, but it often happens that different groups or players get different parts of the protein right. Naturally, the all-hands mode becomes crucial as the key tool for combining the individual discoveries and solutions into the best solution produced by the collective "meta-brain" of foldit players. The original all-hands methodology had at least two key problems:
- the unique solutions quckly got swamped by the singular concensus solution (that is improved by an astounding amount, but still may not have contained some good parts of other solutions),
- there was no clear way to compare sectional goodness of different solutions
- there was no easy way to transfer/incorporate only the portion of the whole structure.
We have addressed these two problems in the recent release, with the following changes:
- We changed the way we cluster the candidate solutions. Now the unique well performing solutions should not be removed from the pool of good candidates. Good solutions that are very similiar to others will no longer be present.
- When viewing other solutions as ghost, their transparency and the coloring scheme points to the specific locations where the ghost solution is better, suggesting the potential partial replacements.
- with the new "copy the secondary sctructure" action you can transer portions of the proteins, and proceed with adjustments and cleanups to produce a better merged solution.
The subsequent all-hands Grand Challenges will be particularly exciting to us, and we hope that they can showcase the power of the collective mind to a much greater extent. As these challenges complete, we will be busily preparing the paper to Nature reporting the ways in which a game can do better than any currently known method for structure prediction.