Constraint Based Approach to the Protein Folding Problem
Abstract
Bioinformatics is a challenging research area where every major
contribution can have significant impact on medicine, agriculture, and
industry.
Among various problems tackled in this area we
are mainly interested in the problem of
the prediction of the spatial shape of a polymer,
given the sequence of monomers constituting it.
This problem can be naturally formalized using constraints over finite domains or intervals of
reals.
The problem
A protein can be seen as a linked list of aminoacids.
There are 20 kinds of aminoacids.
Each spatial, 3D, conformation of the linked list is associated to a given potential.
The problem is that of predicting the spatial conformation that will be reached by the protein that is (should be) the one minimizing the potential.
Two main aproaches: working within a
discrete lattice,
and working
off-lattice.
A survey of our work:
Logic Programming Techniques in Protein Structure Determination: Methodologies and Results.
In Proc. of LPNMR09,
Potsdam, Germany.