Protein Binding as a Virtual World
We who work in VR and it's various sub-fields are often too focussed on achieving nebulous 'realism' to remember one of the cornerstones of VR: Because it is effectively free from the universal laws of our reality (unless its developers say otherwise), a virtual world can effectively be any conceivable environment imaginable.
The VR covered in this article is very much one of the latter kind. The fact that it does not resemble a conventional landscape is not a detriment to its usefulness. In fact this VR is infinitely more useful because of it's nature. It is of course, an interactive 3D simulation of a key protein binding in human cells, which is quite complex, and provides most of the drug resistance for cancer, in the same manner as it works to keep normal cells healthy by sucking out undesirable chemicals.
A view of the VR in operation. In this simulation, everything is colour-coded by it's function. So, helices are purple and blue, ribbons are yellow and coils cyan. At a glance, even whilst the system is moving, it is easy to tell what everything actually is. The large white section is a newly identified inhibitor, and at this point, the simulation is showing how it bonds to one of the pumps two ATP binding sites shown here in dark blue. The user is free to not only watch from any part of the simulation, but interfere with the binding process in real-time by adding other agents into the mix such as chemical compounds and observing, again in real-time, how they affect the outcome.
This use of VR to simulate protein binding at such a complete level, is of course immeasurably superior to the use of static images, and long drawn out physical trials, cancer researchers have been forced to work with in the past. This new environment doesn't replace those methods, but it does supplant them in many instances. Why wait months for the outcome of physical trials on rodents, when you can use the simulation to see which is unlikely to be effective, in a matter of hours? Those the simulation predicts will have an effect can then be run through trials, greatly reducing the expense.
The environment is the brainchild of Southern Methodist University researcher, biochemist John G. Wise. He has combined his simulation with the largest supercomputer the university possesses, in order to give it the raw processing oomph to make it's complex chemical interactions possible.
This is a very different approach than has been used historically in the field of protein structure biochemistry, Wise said. Historically, proteins are very often viewed as static images, even though we know that in reality these proteins move and are dynamic.
The dynamic 3D model already has made it possible for Wise to virtually screen more than 8 million potential drug compounds in the quest to find one that will help stop chemotherapy failure. So far, the supercomputer search has turned up a few hundred drugs that show promise, and Wise and SMU biochemist Pia Vogel have begun testing some of those compounds in their wet lab at SMU.
This has been a good proof-of-principle, said Wise, a research associate professor in the SMU Department of Biological Sciences.
Weve seen that running the compounds through the computational model is an effective way to rapidly and economically screen massive numbers of compounds to find a small number that can then be tested in the wet lab.