Update (9 Dec 2017): Allister Crow has updated his instructions to produce colored AR structures; they are available here.
Scientific publications represent years of work. It’d be nice if somebody read them.
That’s the problem Allister Crow faced as his postdoctoral work was published in early November.
Crow, a structural biologist at the University of Cambridge, UK, was part of a team, led by Vassilis Koronakis, that solved the structure of a bacterial protein called MacB, a pump protein that is involved in antibiotic resistance and toxin secretion. The paper went online November 6 in the Proceedings of the National Academy of Sciences. But how to get people — and especially those outside his immediate field — to notice?
“We’re microbiologists and doing structural biology,” Crow says. “It’s hard to bring that to the general public. There’s quite a distance between us.”
To bridge the gap, Crow created an animated GIF of his protein’s structure and posted it to Twitter, but few of his followers paid attention. Then, he came across an app called Augment, which uses ‘augmented reality’ to overlay virtual objects on the live camera display of a viewer’s smartphone — the same technology that was popularized by the “Pokémon GO” video game.
Augment bills itself as “the first application with which you could visualize 3D models in the real world” — a commercial tool that, among other things, allows users to view 3D projections of design blueprints, or to visualize furniture in the room in which it is to be placed. But there’s nothing to prevent researchers from displaying protein structures, instead.
Crow worked out how to convert his PDB data files into the 3D models Augment uses (you can read the instructions here). He then created an account on Augment, uploaded his file, and associated it with a unique “tracker” — an object that causes the software to project its 3D model into virtual space. Anyone with the Augment app can view that model by pointing their smartphone camera at the tracker — in this case, a printed piece of paper. By rotating the tracker object, the model itself rotates in space.
For his first try, Crow used as a tracker the title page of his PNAS article, tweeting a short video of the effect to his 334 followers. The response was largely positive, he says, and that tweet was retweeted more than 1,000 times and ‘liked’ nearly 2,000 times. But, because the article was not open access (and thus not freely available for download), some people complained. So, he created a second tracker that anybody could use — a black-and-white outline of the structure itself. It’s been downloaded some 500 times.
“I think the augmented reality has captured a lot of peoples’ imaginations beyond those just interested in our work on MacB,” Crow says. “Others think it’s cool, and that’s great to get people interested in structural biology.”
Indeed, according to Crow, it is not structural biologists who are likely to derive much benefit from augmented reality — they can already use the tools of their trade to visualize and manipulate structures on their computers. Rather, it’s the general public — students in classrooms, and families at outreach events — who will see the real payoff. During an undergraduate lecture on antibiotic resistance mechanisms at Cambridge, for instance, Crow distributed printouts of the MacB tracker so the students could view and interact with the protein model at their leisure.
If nothing else, the effect is really slick — a 3D curlicue of blue ribbon floating above the paper, complete with a projected shadow. “I don’t think AR will replace looking on a computer for scientists,” Crow says, “but it’s a great, fun thing to do when teaching undergraduates or trying to engage the general public.”
Proteome GO, anybody?
Jeffrey Perkel is Technology Editor, Nature
1 Dec 2017: This post has been updated to include the PI of the MacB team, Vassilis Koronakis.