A: When I entered grad school, I just
followed my nose and got really
excited by biophysics. There was
this new frontier in biophysics and
a lot of data coming from biologists.
For the first time, we could see
quantitatively how genes interacted.
We had the whole Human Genome
Project and many other types of
large-scale data, and there was a
question of how to make sense of
that data. A lot of people were coming from a lot of disciplines — physics, chemistry, mathematics,
computer science — to integrate
that knowledge and define a new
interdisciplinary field of quantitative biology. I wanted to bring my
training in physics and contribute
to the field that is in its infancy.
A: When you think of cell communication, the most obvious context
might be the cells in our own bodies
because they clearly act as collective
units. But even organisms we think
of as single cells, like bacteria, act
collectively, too. They have very
intricate systems of communicating;
they release molecules in their environment and import those molecules
again so they can tell how many
neighbors they have, for example.
Cell communication is not only
important in organism development,
but it also gets exploited in disease
processes when tissue development
goes wrong such as in cancer. Cell
communication is prevalent and
critical in many organisms. What
can communication do for cells to
help them evolve and survive?
A: In the spring I’ll be teaching
Biophysics II, for graduate students.
Biophysics means a lot of different
things to a lot of people so it will be
a multifaceted course.
A: I like to run because it’s a great
way to explore, especially a new
area. I’ve even run a couple marathons. I also enjoy Ultimate Frisbee,
swimming and hiking.
DEPARTMENT OF COMPUTER
Previous: postdoctoral scholar,
University of California, Berkeley
A: I’m interested in systems secu-
rity and current applications. The
complexity of software is growing
way faster than we can assess or
enforce security properties. It’s
nearly impossible to find and fix all
of these security vulnerabilities. So
my research focuses on protecting
applications even in the presence
of vulnerabilities. It’s more of a
reactive form of security that
detects an ongoing attack and pro-
tects the user’s data and the appli-
cation from harm.
A: Most, if not all, software has
vulnerabilities. To secure our systems, we protect the applications in
several layers that need source code
access. Unfortunately but understandably, most companies only ship
their programs in compiled binary
form. FastBT, a small, lightweight
virtual machine that encapsulates
individual applications, can restrict
access permission and add or
enforce additional security properties on an application on a very
fine-grained level. You don’t have to
set up a full-blown virtual machine
with operating system, disk and
network access; you just push the
application into this small virtual
machine for programs.
Building on fastBT, we’ve implemented Libdetox like “library detox”
that enforces strong security policies on the running applications with
reasonably low overhead.