autoPACK Visualization Challenge

Three-D animators have long sought algorithms that can pack odd-shaped things into tight spaces. Now, Graham Johnson, PhD, a QB3 Faculty Fellow in bioengineering at the University of California, San Francisco together with Arthur Olson’s Molecular Graphics Lab at The Scripps Research Institute (Ludovic Autin, David Goodsell, and Michel Sanner) has created an algorithm called autoPACK that can pack anything into anything—including stuffing molecules into cells to visualize how they interact in space. And, with support from the CG Society of Digital Artists and Autodesk, he is challenging both professional animators and the scientific community to use the algorithm as part of the autoPACK Visualization Challenge (http://autopack.cgsociety.org).

The Challenge provides participants with the necessary ingredients to produce visualizations of HIV in blood serum using open-source models that are as biologically accurate as possible and constantly being updated. The goal: to convey humanity’s complex relationship with HIV in either a short film (under two minutes) or a JPG image.

Johnson hopes the competition will not only build a large community of users and developers who become hooked on the program, but also attract both artists and professional, Hollywood-caliber 3-D animators from outside biology to help build on the open-source project. “They will most want the generic packing algorithm, but our core distribution will always come with the biological applications as part of the GUI,” he says. “And working with DNA and proteins as part of the competition will provide a biological hook into industry that we hope will continue long after the contest is finished.”