Shawn Douglas traces the origins of his current work
fashioning nanoscale DNA robots to his boyhood backyard workshop. From
age five on, he spent countless hours with his father building model
cars, planes, and rockets. “My dad is the most patient and noncritical
person I’ve ever met,” Douglas says. “He never grabbed anything and did
it for me. Without that I don’t think I would be here.”
Despite
the ability he cultivated, Douglas’s desire to study at Yale University
surprised his parents, who didn’t think they could afford it until they
discovered need-based financial aid. After studying literature at Yale,
Douglas eventually opted for computer science, but then realized he
didn’t want a tech job. “I don’t like working on things that are going
to happen anyway without me,” he says. “I had a gut feeling I could take
programming skills into an area that was not mature and make more
impact.”
This thinking steered Douglas to a biophysics PhD at
Harvard Medical School, working with William Shih and George Church. In
mid-2005, shortly after Douglas started working with the synthetic
biologists, Shih heard Caltech’s Paul Rothemund talk about creating
self-assembling DNA shapes and patterns. “I came back to Boston very
excited, and Shawn was the first one with whom I could share the news,”
Shih recalls.
At first Douglas tried to enhance and expand
software Shih had written to design DNA sequences that would
self-assemble. Then he built his own, called caDNAno, from scratch.
Douglas worked on caDNAno through early 2008, limiting his work in the
wet lab and leaving Shih “frustrated with his priorities.” But now
Douglas considers creating the program his most important
accomplishment, and his former Harvard labmate Hendrik Dietz, now at
Technische Universität München in Germany, agrees. “With caDNAno we
could really focus on churning out sequences for all kinds of DNA
objects,” Dietz says.
Dietz made a breakthrough of his own on DNA
self-assembly protocols in late 2007, fuelling a friendly rivalry with
Douglas, with each seeking to trump the other’s group meeting
presentations. That relationship helped spawn two 2009 papers pushing
the field from two-dimensional to three-dimensional assembly and
enabling twisted and curved structures.
Working with Church as a
postdoc, Douglas used his nascent skills in synthetic biology to develop
nanoscale DNA robots. In a 2012 paper, he described DNA boxes loaded
with antibody fragments that could be opened with two different
molecular keys, such as cell surface antigens. “This really inspired
people to dream about the therapeutic potential of
information-processing nanorobots made out of DNA,” stresses Shih.
In 2011, Douglas established an annual competition called BIOMOD to
help students enter the molecular programming community. Every year,
undergraduate teams compete to build the coolest stuff using
biomolecules.
Since Douglas became an assistant professor at the
University of California, San Francisco, in 2013, his lab has worked
towards next-generation DNA nanorobots. “The Science paper from 2012 was
a prototype,” he emphasizes. “Something that will be used in patients
will probably look very different.”
|
