Protein folding, simulation vs. experiment

Prigozhin et al. has brought simulations and experiments of protein dynamics closer together than ever before. They do this by creating a clever new protein that gives more detailed experimental insight and simulations on a state-of-the-art molecular dynamics supercomputer. Their insights add another brick into the path that scientists have been building for decades to bridge protein experiments and simulations. The only way to do this are through advances like this paper - creating experiments that are faster than ever before while creating simulations that can simulate reality for longer than ever before.

For decades researches have been trying to simulate the inner workings of biology from the bottom up. One of the smallest pieces of this giant puzzle is the protein - usually consisting of a few thousand atoms. However, even for a few thousand atoms the best computers in the world can only simulate them for one millionth of a second! That’s a problem and its one that Maxim Prigozhin, Shu-Han Chao, Shahar Sukenik, Taras Pogorelov, and Martin Gruebele have worked on in their latest work and partially solved using a unique protein construct and some stellar simulations.

Biologists who want to validate their simulations. More opportunities are emerging to connect experimental and simulated data which will help guide new experiments and revise all energy functions that are used for simulations.

Prigozhin et al. take advantage of the fact that they can add in fluorescent residues into their protein, without disturbing its structure. The fluorescence tells them how close certain residues are, and they can monitor this at a very fine timescale (microseconds). At the same time they use a supercomputer that can reach 0.1 microseconds of simulation. They get qualitative agreement using these methods which is a validation of simulations and comforting to theorists all around that their somewhat adhoc molecular dynamic schemes truly do approximate reality.

This study is only on one protein - and its a small protein compared to the rest of the proteome which have huge proteins!. Also, this one protein has a very special structure that allows it to be mutated with different special fluorescent residues which in turn allows scientists to target it with the most advanced experimental technologies. There are very proteins like this one, so this trick will only work once (maybe twice). Still, it is a steady step in the right direction - like all science!