|
|
|
 |
Research highlights:
*** This paper
should be of interest to the
following communities and
disciplines.
Maths
/ Physics / Computer
science (Network
science and graph
theory):
A mathematically rigorous
method to compare any two
networks (independent of
origin, function or kind),
with overlapping nodes.
This is done by
introducing a measure
called Shortest Path
Alteration Fraction
(SPAF), technical details
regarding which appear in
Theorem 1. SPAF=1 leads us
to all shortest paths
present in either network
but not both.
General
biology:
This approach immediately
presents a fresh and
general framework to
address mutations from the
level of a single protein
to an entire organism. In
fact, the latter has the
potential to deliver
systems-level
perspectives. We
computationally explore
phenotypic alterations
arising from
mutations in five
microbes, each from a
different taxon. Effects
reported in literature are
recovered and new
predictions made.
Differences between a pair
of similar networks at any
level, including say at
the ecological level can
be explored by this
method.
Microbiology:
Phage resistance in
mycobacteria is
significant in its own
right due to the emergence
of multidrug resistant
strains and the promise of
phage therapy. We
scrutinise the
effectiveness
of our
procedure
through
extensive
theoretical
and
experimental
tests in such a
system.
A
mutant of Mycobacterium
smegmatis mc2155
-- resistant to
mycobacteriophage D29 -- is
generated
and is characterised with
significant phenotypic
alterations. Whole-genome
sequencing identifies
mutations, which cannot
readily explain the
observed phenotypes. We
show the utility of SPAF
towards mapping
the present
genotype-phenotype
relation.
***
Networks with a scale-free
degree distribution are
widely thought to promote
cooperation in various
games. We demonstrate that
this need not necessarily
be true. For the same
degree distribution and
indeed the very same
degree sequence, we
present a variety of
possible behaviour. We
also reevaluate the
dependence of cooperation
on network clustering and
assortativity.
***
Experimental validation of
predictions from the analyses of
Protein Contact Networks or
Residue Interaction Graphs is
scarce in literature. Our
collaborators from Devrani Mitra
Lab have directly verified
the results predicted earlier by
us in Bioinformatics
[Oxford] 31 3608-3616 (2015).
***A
Systems Biology approach consisting of
an iterative cycle of theoretical
techniques (delay differential
equations and Monte Carlo simulations), coupled
with experiments; which
presents new evidence
for secondary mechanisms of host
lethality in Mycobacteriophage-Mycobacterial
host
interaction.
*** A
fresh, general and simple method of
feature extraction (image processing),
using networks which can be implemented
in form of a smart, fast and portable
device. Applications of such a fluctutation
based system and method have
been demonstrated in non-invasive
diagnostics and biometrics.
Patent
Filing no. IPO 628/KOL/2015
***
A new edge-based metric which
demonstrates slow-poisoning in
networks and shows how edges and
nodes, which are important for
infrastructure, biological and other
networks, can be identified.
Physical
Review E 91 022807
(2015)
*** Initiation
of a framework towards understanding
light-dark transition in photoreceptors. Our
approach of introducing
differential networks and identifying
important residues: (1)
minimises extensive photo-cycle kinetics
procedures, and, (2) is helpful in
providing first-hand information on the
fundamentals of photo-adaptation and
rational design of photoreceptors in synthetic
biology inspired structural
optogenetics.
Bioinformatics
[Oxford] 31
3608-3616 (2015)
Postdoctoral work:
***
The first work in complex network
literature outlining how and why
multiple metrics (and higher moments of
metrics) should be measured
simultaneously in networks and
procedures to identify informative or
redundant metrics. The usefulness
of this approach is demonstrated in
complex systems with examples from
systems biology.
|
|
