How Virtual Reality Can Help Scientists, Such As Yourself, Better Understand Molecules

Author: Jessie MNGL Suzuki, almost Ph.D. In part one, Jessie talks about using virtual reality in the lab. Next week in part two, he’ll show you how to build your own VR setup.

Have you ever visited a “new” place only to realize that you have actually been there before? You go try out a new beach, but as soon as you get there, oh that shoreline, that cliff, that seafood shack, your grandmother took you here once as a child! Or, have you ever packed a suitcase or trunk by looking around for the next object which can fit perfectly into an available open space? Ah, the satisfaction of a fully Tetris-ified trunk. Or, can you accurately throw a ball to hit a distant moving target? Ok, me neither, but some people can! All of these abilities rely on the human brain’s incredible capacity to understand, remember, and manipulate 3-dimensional physical spaces. You can harness that same passive, intuitive, and sophisticated mental ability to help advance your understanding of the molecules you study, using virtual reality (VR).

                        In 2017, I was a PhD candidate at University of California, Santa Cruz in the Center for Molecular Biology of RNA. I was visiting an old artist friend who is a visual machine-learning developer. This friend put me into a VR drawing program. While the rest of the gathering sat around watching what I was doing on the big screen TV, I was whisked away to a sparse desert landscape with a magic paintbrush that could deposit bold 3-dimensional lines of ink in the air, wherever I placed them. I created a sculpture portrait of my friend’s shih tzu, Dolma. I drew a campfire, a little house, a human brain. I walked around these astonishing objects that I had created. I stepped inside the brain and saw the little cave, the convoluted backsides of all my brushstrokes; I suddenly gasped and yanked off the headset, demanding to know: Could I view a PDB file in here?

The Protein Data Bank (PDB) currently houses over 185,000 files representing biological molecular structures, deposited by teams of scientists from around the world (1). These 3-dimensional structures, obtained at great cost, provide information about proteins, nucleic acids, and small molecules that inform basic and translational scientists, and they are free for anyone to use. Billions of dollars and uncountable painstaking work hours have gone into the production of this treasure trove of 3D models, whose beauty and utility are enfeebled by the common practice of reducing them to 2-dimensions for viewing on paper or a screen. To see into the heart of one of these models, we slice the molecule open, creating an artificial planar cross section. We capture an interior view but lose critical context and connections. Unlike genetic sequence information, which loses little in the conversion from quaternary ATCG code in a cell to binary ones and zeros in a computer, the loss of depth from the 3-D models to the screen or paper representation represents a major and tragic information degradation.

I spent years fervently proselytizing the power of VR for molecular viewing in the halls of my university. Initially, I had to walk down through redwood forest (I know, poor me, but it’s far) to the vast concrete basement of the campus library to use the VR rig used by students in the video game design major. I helped author a grant proposal to buy a VR for the Science and Engineering Library, but VR was dismissed as a gimmick, at best – a way to wake up undergrads numbed by technology. The grant was rejected. I appealed to my department, also denied. Gradually, I cajoled colleagues to make the trek across campus to join me on my adventures. They found insights, returned, told other people, and started bringing their labs. The first time Dr. Manny Ares shared physical space with the molecular structure he has spent a career elucidating he said, “I have been studying this stem-loop for more years than it has nucleotides, and this is the first time I’ve ever actually seen where it lives.” My own Ph.D. advisor, Dr. Alan Zahler, is not quite a luddite, but neither is he an early adopter; my paper was the first he was ever willing to put up on this new BioRχiv thing. He finally made the trek down to the library where he spent hours in structure after structure. When the battery power on the hand controllers finally ran out, he took off the headset and said, “Ok, I get it now, this is important.”

A midday walk in the woods can be lovely, but its inconvenient to have to go across a large campus to use a piece of lab equipment. You wouldn’t want to walk that far every time you need a whiteboard. It was Dr. Ares, who finally put his money where all our minds were. After yet another funding lead dried up, he asked, “How much would this even cost?” and found the money. He purchased a top-of-the-line VR set-up using funds donated by a generous former undergraduate researcher who eventually succeeded in biotech and wanted to show support for the lab. These days, when a new molecular structure is published, or a genetic screen identifies an important residue, I just pop into the conference room down the hall from my laboratory to enter virtual reality. Now, visiting seminar speakers get a half hour visit with their proteins of interest, “Hey nice talk today! So, you ready to finally meet a toll-like receptor in person?” These visitors often leave with a list of insights, residues they want to mutate, new theories, and a clear memory of the space their molecules occupy – like a beach they visited once and now will always remember. I occasionally get a message from a frustrated grad student, back at the visiting professor’s home institution, who is now fielding their P.I.’s urgent requests to set up VR. One told me, “Just because I’m a bioinformatician doesn’t mean I know how to set up VR!” Fear not, my frantic friend, I wrote this document for you! This is how you do it.

You will need a space to play, a computer, an HTC Vive VR headset, and some free software! Our humble set-up down the hall is top-of-the-line and cost us about $4,000.

Acknowledgements

Manny Ares, Scott Seiwert, Al Zahler, Susan Strome, Aleta Dunne, UCSC, UCSF

References

(1)        RCSB Protein Data Bank: powerful new tools for exploring 3D structures of biological macromolecules for basic and applied research and education in fundamental biology, biomedicine, biotechnology, bioengineering and energy sciences. Burley et al., (2021) Nucleic Acids Research  doi: 10.1093/nar/gkaa1038

(2)        Virtual-reality applications give science a new dimension. Matthews D. Nature. (2018) doi: 10.1038/d41586-018-04997-2.

(3)        UCSF ChimeraX technical support website https://vr.ucsf.edu/

(4)        A Genetic Screen in C. elegans Reveals Roles for KIN17 and PRCC in Maintaining 5’ Splice Site Identity. Suzuki et al., PLoSGenetics (2022) doi: TBD