By Rachel Tompa, Ph.D. / Allen Institute

Dan Toloudis has spent his career designing graphics tools for computer animators. In previous graphics engineering jobs, he wrote programs that enabled artists to create lifelike dragons at DreamWorks Animation or spinning cars for a digital Ford commercial.

Now he spends his days thinking about human cells.

Toloudis is one of the engineers behind AGAVE, a new graphics tool developed at the Allen Institute for Cell Science to help researchers visualize the delicate and dynamic inner workings of human stem cells.

AGAVE uses path trace rendering, a graphics technique straight from Hollywood’s animation studios that is shedding new light — literally — on the microscopic structures inside our bodies.

Path tracing is responsible for the realism of the dancing, crashing waves in Disney’s Moana, or the rich shadows in caves and hallways in the recently announced release of Minecraft. The technique is all about light and dark, using sophisticated graphics techniques to create a 2D image on a movie or computer screen that matches what we expect from the 3D world around us.

Light and shadow for scientific discovery

“Think of the way light reflects and refracts inside a volume. With something like water, there are a lot of subtle secondary effects from how light bounces around,” Toloudis said. “It’s all about understanding the physics of that effect.”

As visual creatures, we instinctively rely on those subtle effects to understand and navigate our 3D worlds. Path tracing, also known as ray tracing, is a method that captures rays of light in a scene as they bounce from a light source off of different surfaces in the field of view, finally making their way to the camera. Pictures or animated movies that recreate realistic light and shadow look more natural to us, more understandable, because of these complex layers of reflections and shadows.

The same is true for scientific images of cells. Researchers can better interpret images of cells if the light and shadows on the data representations of cellular structures match what our brains expect from natural light. This visualization method can thus help scientists understand the spatial relationships between different parts of the cell, relationships that may be crucial to understanding how our cells work.

“This technique is letting us interpret images in new ways,” said Graham Johnson, Ph.D., Director of the Animated Cell team at the Allen Institute for Cell Science, who also worked on AGAVE. “Path tracing has been a major challenge to do on a standard computer for the last several decades. But now, in near real-time, we have tools that are giving us a more intuitive understanding of where structures exist relative to each other in our cells.”

The visualization tool combines path tracing with another graphics method that generates 2D images of 3D objects (such as cells), known as volume rendering. Path tracing has been used in Hollywood for over a decade, and several existing medical and scientific imaging methods rely on both path tracing and volume rendering.

But it’s only relatively recently that graphics processing units, or GPUs, of personal computers and gaming systems have gotten large enough to handle these power-hungry techniques. The first video games that rely heavily or completely on path tracing are only just coming on the scene. That’s why it was the right moment to adapt the technique for cell biology too. AGAVE won’t work on all computers, but most modern laptops now have the capabilities needed to run the program.

The program is available as a beta release on allencell.org in both a full, downloadable application, and a simplified browser version. Toloudis and the other engineers are currently working on several new features to add to the next release of AGAVE, including the ability to apply the tool to real-time movies of cells.