Back Me Up Puzzle

Started by beta_helix

beta_helix Staff Lv 1

Over the past few decades, enzymes have found their way into the spotlight for performing difficult chemical reactions in the laboratory and industrial settings. This interest is driven by their ability to efficiently catalyze reactions in water at room temperature with incredible efficiencies (rate enhancements of up to 1019). In addition enzymes have stringent specificities for the substrates they take in and the products they put out. While over 280,000 enzymes have been currently identified in nature, there are still many reactions that researches commonly use in the laboratory for which there is no known enzyme to catalyze the reaction. One such reaction is the Diels-Alder reaction. The reaction was discovered in 1928 and won a Nobel prize in 1950. Since then it has become a cornerstone reaction for organic chemistry and can be found in every introductory organic chemistry text as well as many patents in the pharmaceuticals, materials, fine chemicals, and agrochemical industries.

In this puzzle we present an enzyme we have designed that is able to catalyze the Diels-Alder reaction. Currently it has low levels of activity but we believe that by increasing the number of contacts to the ligand we can increase its activity. To do this we have engineered a partial helix that has shown some increase in activity, but we believe YOU CAN MAKE IT BETTER! The goal of this puzzle is to stabilize this helix and make a more ordered secondary structure motif (e.g. Helix-Loop-Helix). Your results will be selected based on the lowest energy and best secondary structures. The top designs will be tested in the laboratory for their ability to catalyze this reaction.

John McLeod Lv 1

I know very little about amino acids, but have hit on a method for making some progress (after some days of frustration that I think some other folders have been sharing on this one).

Simply put, I attached two or three bands to the structure and then compressed it against the frozen main part of the protein, then disabled the bands and wiggled it back up, interrupting the process for mutates and shakes. (this was after adding a few additional amino acids to the yellow bit) Apparently it's known that doing this compression cycle and then "letting it breath" will sometimes help mutate discover better combinations, and that fiddling with "clash importance" under "behavior" will sometimes also result in progress in somewhat the same way.

If this results in a few Rank Downs as others try this I'll still consider my time well spent if someone can point to where I can learn a bit more about how those amino acid patterns are supposed to look.

beta_helix Staff Lv 1

From this puzzle we got some very nice, natural looking secondary structure elements designed by Foldit players.

We are building some proteins based on player models and will be testing them in vitro over the next few months!

It's very exciting but unfortunately it takes a while to build and purify proteins so we don’t yet know what worked and what didn't.

As soon as we find out we will let you know in a new blog entry (so you don't have to keep checking this one).