DIMACS Workshop on Combinatorial Clustering and Multi-Domain Protein
Structure Analysis
June 26-27, 1998
DIMACS Center, CoRE Building (Room 431), Rutgers University, Busch Campus,
Piscataway, NJ
URL: http://dimacs.rutgers.edu/Workshops/CombCluster/announcement.html
Principal Organizers:
* Sylvia Spengler, Lawrence Berkeley NationalLaboratory
* Manfred D. Zorn, Lawrence Berkeley National Laboratory
* Inna Dubchak, Lawrence Berkeley National Laboratory
* Fred Roberts, Rutgers University, DIMACS
* Ilya Muchnik, Rutgers University, DIMACS
* Casimir A. Kulikowski, Rutgers University, CS Department
Presented under the auspices of the Special Year on Massive Data Sets and
the Special Year in Mathematical Support for Molecular Biology.
----------------------------------------------------------------------------
The combinatorics, graph theory and algorithms branches of discrete
mathematics are powerful techniques for use in analysis of diverse sets of
data. Combinatorics introduces clustering models, forms a basis for the
interpretation of results, and brings new ideas to clustering methods. The
area of combinatorial clustering provides new opportunities for systematic
(routine) and comprehensive study of very large databases containing highly
complex non-regular elements. However, researchers developing methods for
combinatorial clustering often do not collaborate with specialists in the
creation, management and analysis of large, complex databases.
The goal of this workshop is to foster and facilitate such a collaboration
by focusing on a particular application. The specific problem we wish to
address is in the area of protein structure and analysis of multi-domain
proteins (as an example for other type of interactions among large
bio-molecules). By hosting this workshop, our aim is not only to stimulate
research at the interface between these disciplines, but also stimulate
development of ideas and methods in each field.
We plan to have two type of sessions: theoretical, devoted to mathematical
models and clustering procedures, and practical, concentrated on
sequence-structure relations in biomolecular systems, mostly on multi-domain
protein analysis.
Following applications of clustering to the multi-domain protein analysis
will be discussed among others:
* amino acid sequence segmentation which determines domain structures
of proteins,
* automatic procedure to collect libraries of protein domains,
* contact map method for protein 3D-structure analysis and protein fold
classification,
* inter-domain contact aggregates,
* correlation between contact structures and sequence segmentation.
URL: http://dimacs.rutgers.edu/Workshops/CombCluster/announcement.html
