Paper abstracts
- Search and Rescue: logic and visualisation of
biochemical networks
Nicos Angelopoulos, Paul Shannon and Lodewyk
Wessels
We utilise a recently introduced Prolog package that
allows communication with the R statistical software to develop a
graph-centric suit of procedures that allow the exploration of
biological data and hit-lists from within Prolog. We show how a number
of public protein-protein interaction databases can be intuitively be
represented as facts before we present search algorithms that are
naturally expressed in logical terms. Visualisation of the resulting
graphs is done via low-level communication to Bioconductor's
RCytoscape package. We illustrate the utility of the integrative
Prolog platform using two public databases and a graph search to
connect elements of genes involved in cell motility.
- Protein Loop Modeling via Constraints and
Fragment Assembly
Federico Campeotto, Alessandro Dal Palu,
Agostino Dovier, Ferdinando Fioretto and Enrico Pontelli
Methods to predict the structure of a protein often rely on the
knowledge of macro-sub-structures and their exact or approximate
relative positions in space. The parts connecting these sub-structures
are called loops and, in general, they are characterized by a high
degree of freedom. The modeling of loops is, thus, a critical problem
in predicting protein conformations that are biologically
realistic. This paper introduces a constraint that models a general
multi-body system, and shows its application to the protein loop
modeling, based on fragments assembly, along with filtering
techniques, inspired by inverse kinematics, that can drastically
reduce the search space of potential conformations.
- Qualitative Models of Cell Population Dynamics as
Constraint Satisfaction Problems
Tom Kelsey and Steve
Linton
Existing approaches to the modelling of mammalian
ovarian cell dynamics involve differential equations. This
methodology has several defects, including the inclusion of modelling
assumptions that may not be supported by empirical data. We present
the derivation of Constraint Satisfaction Problems based on
qualitative properties of the underlying cell dynamics. When combined
with boundary conditions, the solutions provide a range of models that
can be assessed for validity using residual errors from empirical
observations.
- Atom Mapping with Constraint Programming
Martin Mann, Heinz Ekker, Peter F. Stadler and Christoph
Flamm
The mass flow in a chemical reaction network is
determined by the propagation of atoms from educt to product molecules
within each of the constituent chemical reactions. The Atom Mapping
Problem for a given chemical reaction is the computational task of
determining the correspondences of the atoms between educt and product
molecules. We propose here a Constraint Programming approach to
identify atom mappings for ``elementary'' reactions. These feature a
cyclic imaginary transition state (ITS) imposing an additional strong
constraint on the bijection between educt and product atoms. The
ongoing work presented here identifies only chemically feasible ITSs
by integrating the cyclic structure of the chemical transformation
into the search.
- An ASP Implementation of PhyloWS
Enrico
Pontelli, Tran Cao Son, Tiep Le and Ngoc-Hieu Nguyen
This
paper illustrates the use of ASP technology in implementing the
PhyloWS web service API|a recently proposed and community-agreed
standard API to enable uniform access and interoperation among
phylogenetic applications and repositories. To date, only very
incomplete implementations of PhyloWS have been realized; this paper
demonstrates how ASP provides an ideal technology to support a more
comprehensive realization of PhyloWS on a repository of
semantically-described phylogenetic studies. The paper also presents a
challenge for the developers of ASP-solvers.
|