Many processes in biology are so complex and hard to model
using physics laws. The problem of how a Drosophila egg manages
to mix its contents would naively appear to be in this category. However as it
turns out, it can be understood very well with a physical model that only
has a few ingredients: a fluid, semiflexible fibers, and motors that
walk on them. Using hydrodynamics, and nonlinear equations,
many intriguing aspects of this system can be predicted and found
to be in excellent agreement with observation. The long range nature
of hydrodynamic flows allows a relatively small number of motors
to drive bulk mixing of the cell. The mixing is aided by chaotic
motion of microtubules, and this motion can be well understood
by modeling the motion of semiflexible chains with kinesin
motors walking up them.
Entropy for isolated systems
How do the laws of statistical mechanics
emerge from quantum mechanics? Although there is a pretty good
explanation classically of this question, the
quantum case is certainly not as clear. Earlier work by the author and
others proposed the
Eigenstate Thermalization Hypothesis.
More recent work by the author has examined a more subtle
problem, to understand the thermodynamic entropy for
an isolated quantum system.
- Using simulations techniques borrowed from
quantum and statistical physics,
masses of data from microarrays can be analyzed
to determine the most important genes in
diagnosing similar cancers. For example, less than 15 genes
can be used to perfectly predict which of four related
childhood cancers a patient has.
- By understanding the underlying physics of
these devices, their reliability can be greatly
Polymers in a vacuum
The 2002 Nobel prize in Chemistry was awarded in large
part for the discovery of how to put single protein
molecules into a vacuum. This is then used to do mass
spectrometry on these large molecules such as proteins. Despite this
technology`s great importance, very little had
been done to understand the behavior of polymers
in this situation. Conservation of angular momentum
alters the statistical properties, and in many circumstances,
chains can show oscillatory behavior very different from
what is found in solution.
Dynamics of magnets
Nonequilibrium dynamics of magnetic systems
have a wide variety of
applications including to hard drives.
New phenomena explored are:
Often avalanches should be viewed
more like combustion than previously understood, where
the flipping of magnetic spins was thought to be more like the
the falling of dominoes.
Hysteresis loop "multicycles"
Domain patterns seen
going from positive to negative applied
field are not simply related to what happens when
the field is inverted.
One dimensional heat conduction
The dynamics of one dimensional system, such
as nanotubes, show anomalous dynamics because
of conservation laws. One of the most
striking consequences of this is that the
heat transfered across such a system does not
scale with the inverse of the system length, but
instead a power of it. Determining the exact
relationship is challenging and requires the
use of highly nonlinear models. By devising a new
model incorporating random collisions but still
obeying conservation of energy and momentum, a clearer
understanding emerged of the large scale dynamical