Last week I attended VIZBI 2013, the 4th international conference on visualizing biological data. I was wowed by the variety of visualization techniques and tools presented, as well as by the high quality of art and design I saw on display . The conference covered a diverse range of visualization problems, from pure data visualizations of genomic, expression and even epidemiological data, to renderings of biological structures at various scales.
One of the posters presented at the conference that really impressed me showed off the entries for a contest, just finished, to render artistic visualizations of HIV in blood serum.
This contest was partly motivated by the desire of the team behind the autoPACK project to make the biological art community aware of the incredible technical resource that they are developing. Simply put, autoPACK allows an artist to take a variety of 3D shapes (such as balls, cylinders, puppies, model cars or, more relevant to biological visualization, proteins and lipids) and pack them together as closely as possible, without touching, within a specified volume or over a given surface within their 3D modeling environment of choice. When visualizing biological structures, the lipids and proteins that make up our cells can thus be virtually “packed” together into convincing membrane structures by this software, thus creating highly realistic simulations of cells and other biological structures without much human intervention.
One of these biological structures that the autoPACK group developed is the HIV virus model. Nowhere near as simple as I naively expected, the virus is composed of an outer lipid membrane, studded unevenly with protein “spikes”, that holds within it various proteins essential to its functioning inside the cell, as well as a “capsid”, a capsule made of protein that in turn holds all the viral RNA inside. The authors of autoPACK provided a very realistic model of this viral structure for use in a variety of commonly used 3D modeling programs, and asked the community of medical illustrators, biologists, artists and enthusiasts to put together a still image or video that conveys this structural data in some artistic fashion.
The community rose splendidly to the challenge: The 88 entries, a mix of still image and videos, can all be seen here.
Looking through the various entries, I was struck by the wide variety of artistic styles used. From “realistic” renders (albeit employing some great lighting) such as this, or this, to highly metaphorical ones, like these two.
There was even some very imaginative use of Lego bricks.
The above are offered simply as examples of the sheer diversity of entries. There are, to my eye, plenty of outstanding works of art to choose from in this collection, and taken together they hint at the dizzying possibilities now available in medical and scientific illustration.
I had a couple of other, perhaps obvious thoughts about this contest:
First, what could only be done ten years ago by a team of specialized medical illustrators loaded with tens of thousands of dollars worth of tools, is now becoming available to the average artist, even hobbyist. As techniques for quickly acquiring and visualizing biological data – such as autoPACK – become accessible to casual users, as increasingly sophisticated visual effects become an everyday part of the artist’s toolkit, and as the hardware required to run 3D programs becomes cheap and ubiquitous, the kinds of visualizations that we’re used to seeing from major movie studios will become commonplace.
Second, this contest is a truly successful example of crowdsourcing in the name of citizen science. The challenge was by no means easy, and yet the consistently high quality of the results, not to mention the amount of work put into just about each entry, pointed to a highly motivated community. And no wonder: The ultimate cause, to help rid humanity of the scourge of AIDS, is being supported here by the donated talent, ingenuity and time of the contestants. The rules of the contest were such that artistic freedom was if anything boosted by the restrictions, which amounted to using the model as a starting point, and a request not to use otherwise copyrighted material in the final work.