Living matter in action, at the
smallest length scale epitomized by
biological motors, operates in
nonequilibrium steady states (NESS).
In NESS, the detailed balance is
broken such that nonvanishing energy
and material currents are constantly
fed and flow out of the system. Together
with the microscopic underpinnings
underlying the functions of
individual motors, we are interested
in quantifying how energy,
information, and material balance in
biological systems contributes to
the emergence of cellular
organizations.
Structure,
dynamics, and function of human
chromosomes
The packaging of chromosomes,
giant chain molecules made of
hundreds of megabase-long DNA,
into a small volume is truly
remarkable. Recent advances in
Hi-C and imaging technique have
ushered in a new era of research
on genome organization, which is
bound to reveal the origin of
the cell type-dependent gene
expression in due course.
Despite these advances, our
understanding of the dynamics of
chromosome/genome at varying
scales of space and time is
still in its infancy.
Seemingly daunting problem at
a first glance, polymer
physics idea and molecular
simulations provide glimpses
into the link between the spatiotemporal
dynamics of chromosome and its
characteristic chain
organization. We try to explore
the effect of physical/energetic
constraints in chromosome
architecture on the dynamics,
and learn the physical basis of
gene expression.
Olfaction
The chemical
space of odors is vast, yet when
odorant molecules interact with a
finite set of receptors, the brain
projects this information into a more
manageable perceptual space. This
process involves a significant
reduction in dimensionality.
Compressed Sensing (CS) is a powerful
algorithm that can reconstruct
high-dimensional signals from data
compressed into lower dimensions,
particularly when the signal is
sparsely represented. Our analysis of
recent Drosophila connectomics
data reveals that the Drosophila
olfactory system is well-suited for
CS. We demonstrate that the neural
activity of projection neurons (PNs)
can be accurately recovered from the
lower-dimensional responses of
mushroom body output neurons (MBONs).
This reconstruction is supported by
electrophysiological recordings across
a wide range of odorants. By analyzing
the residuals between measured and
predicted MBON responses, we have
visualized the perceptual odor space
and examined how different odors can
be distinguished from one another. The
study highlights the importance of
sparse coding in the receptor space
for odor identifiability, shedding
light on how odor perception varies
with concentration. Additionally, when
the olfactory system is exposed to
multiple odorants simultaneously, the
neural activity profile of PNs becomes
saturated. This saturation impairs
signal recovery, leading to a
perceptual state similar to "olfactory
white." Our application of CS to Drosophila
connectomics data offers new
quantitative insights into how odors
are represented within the fly brain,
deepening our understanding of
olfactory processing.
Origin
of
heterogeneous
and slow
dynamics in
biomolecular
systems
Single
molecule time
trajectories
of
biomolecules
provide
glimpses into
complex
folding
landscapes
that are
difficult to
visualize
using
conventional
ensemble
measurements.
Recent
experiments
using single
molecule measurements
and
theoretical
analyses have
highlighted
dynamic
disorder in
certain
classes of
biomolecules,
whose dynamic
pattern of
conformational
transitions is
affected by
slower
transitions
between hidden
internal
states. In
recent years,
we have not
only
articulated
the existence
of dynamical
heterogeneity
in Holliday
junction
dynamics
(Nature Chem.
(2012))
and in kinesin
function
(Traffic
(2017)), but
also developed
theories
to extract the
extent of
heterogeneity
using
a single
molecule force
spectroscopy (Phys.
Rev. Lett.
(2014), PNAS
(2017)).
The
issues of
properly
analyzing
dynamical
heterogeneity
in the
dynamics of
biomolecules
and
deciphering
microscopic
underpinnings
of the
heterogeneity
are one
of the main
research
themes of the
group.
Allostery
in biomolecules
Allostery
refers to a long-range communication
between two remote sites in biomolceuls.
Key question concerning allostery would
be: For a given conformational ensemble
of a protein, which are the key residue
for allostery? What are the allosteric
signaling pathways? Investigating a
variety of structural ensemble of
protein with and without cognate ligand
we aim to decipher the physical basis of
allostery.
Contact
Information : Changbong Hyeon, Professor,
School of Computational Sciences, Korea
Institute for Advanced Study, Seoul 02455,
Korea +82-2-958-3810
(tel)
