|
|
Talks
Erida Gjini (Instituto
Gulbenkian de Ciencia, Oeiras, Portugal)
" Integrating antimicrobial treatment
with host immunity in resistance
management"
Abstract: The evolution and spread
of antimicrobial resistance is a major
global problem, and a cause of substantial
human mortality. As the discovery of new
antibiotics does not follow the rate at
which new resistances develop, a more
prudent use of available drugs is needed.
Here I present a mathematical modelof
within-host infection dynamics that
combines the effects of pathogen clearance
by the host immune system and by the
antibiotics. Computer simulations and
mathematical analysis are used to evaluate
treatment protocols in order to identify
those that can restore patient health and
limit the overall load of bacteria and
selection of resistance. The study focuses
on infections with pre-existing
resistance, and explores two main
treatment strategies: the classical
treatment, characterized by fixed drug
dose and treatment duration, and the
adaptive treatment that closely follows
infection outcomes and patient symptoms.
Our results highlight treatment strategies
that promote strong synergy between host
immunity and the antimicrobial drug. This
can be achieved by moderate treatments
that combine appropriate timing of
treatment onset, reduced drug dosage, and
short treatment durations. The model is
developed for bacterial infections but our
framework and findings may apply to other
medical scenarios featuring drug
resistance.
Rui Borges (FCUP) " Detection
of CNVs in BRAC1 and BRAC2 genes – a
hierarchical Bayesian modeling approach"
Abstract: Copy
number variation (CNV) of DNA fragments is
a particular type of structural genetic
variation (Feuk, Marshall, Wintle, &
Scherer, 2006). The characterization and
identification of CNVs has been an
instrumental tool to understand different
diseases. As an example, gene copy number
alterations are frequent in cancer cells
(Cappuzzo et al., 2005).
In this study, we present a Bayesian
inferential framework to detect and
identify CNV in protein coding genes.
Using the coverage readings from next
generation sequencing methodologies, we
pre-transform the study individual
coverage, calculating the ratio between
the individual and a standard coverage
profile. A hierarchical model is then used
to assess the posterior ratio distribution
for each DNA fragment, from which
structural information about the relative
number of copies can be easily retrieved.
We implemented a robust Bayesian
hierarchical model, with the major
advantage of including the degrees of
freedom parameter, which controls the tail
behavior of the DNA fragments coverage
means distribution. The implementation of
the robust model was done using Markov
Chain Monte Carlo (MCMC) methods. Finally,
we present and discuss two case studies
with the BRCA1 and BRCA2 genes.
Md. Haider Ali Biswas
(Mathematics Discipline, Science Engineering
and Technology School, Khulna, Bangladesh)
" Transmission Model for Nipah Virus
(NiV) Infection"
Abstract: Nipah
virus (NiV) is a member of the genus
Henipavirus which is a new class of virus in
the Paramyxoviridae family. It is
transmitted to human from infected fruit
bats. A single species of fruit bats of the
genus Peteropus, P. giganteus, lives in
Bangladesh and is widely distributed
throughout the country. Recently, Nipah is
detected as a viral zoonotic disease caused
by Nipah virus with high mortality rate (an
estimated 77%, 2001-14; 100%, 2011-12) in
Bangladesh, which is a gigantic threats to
the public health of Bangladesh. In this
paper, we introduce a mathematical model to
discuss the behavior of the transmission of
Nipah virus (NiV) from bats to human in
terms of ordinary differential equations
(ODEs). The behavior of the dynamics of NiV
transmission has been illustrated by the
numerical simulations.
Sponsors
|
|
|
|