Currently, there are four main projects under the scope of the
Newcastle Bioinformatics Initiative. They deal with the identification
of new problems, high performance computing, Single Nucleotide Polymorphism data analysis and Metaheuristics.
Identification of Novel Optimization Problems in Bioinformatics
Project leaders:
M. Fellows, P. Moscato and K. Weihe
Personnel involved:
R. Berretta, M. Fellows, H. Fernau, G. Lancia, P. Moscato, E. Prieto, F. Rosamond and K. Weihe
Description: Many problems in Bioinformatics may be primarily classified as either having an efficient algorithm (i.e. polynomial-time) or being
NP-hard (for which it is conjectured that no efficient algorithm exists). Additionally, the field of Bioinformatics and Computational
Molecular Biology is characterized by massive amounts of data. As a consequence, NP-hard problems were regarded as “intractable”, and
only polynomial-time algorithms were considered. This greatly restricted the scope of problems that were addressed. More recent opinion
believes that the use of “well-engineered” heuristics and metaheuristics could achieve new breakthrough discoveries towards the solution
of NP-hard problems.
Distributed Computing for Bioinformatics
Project leaders:
P. Moscato
Personnel involved:
C. Cotta, A. Mendes and P. Moscato
Description: This subproject involves several sub-goals that are being developed as concurrent directions of action. Uncover new
problem of relevance as well as other challenging large-scale Optimisation problems arising in Bioinformatics that would require HPDC
algorithms; Formalize new problems as combinatorial, non-linear, mixed or multi-objective Optimisation problems; Identify the best way
of addressing and solving these problems using metaheuristic methods based on academically available HPDC systems and, when justified,
hybridise the methods with exact algorithms; Implement in HPDC systems the new powerful algorithms based on these techniques.
Algorithmics for Single Nucleotide Polymorphism (SNP) Data Analysis
Project leaders:
P. Moscato and R. Berretta
Personnel involved:
G. Lancia, A. Mendes, P. Moscato and R. Berretta
Description: Newcastle is a very important centre for SNP analysis. One of NBI’s collaborators, Prof. Rodney Scott, is a Co-Director
of the The Clive and Vera Ramaciotti Centre for Gene Function Analysis, which is a collaborative effort on the part of five NSW universities
and the major medical research centres in NSW. Through this collaboration we plan to address the application of novel algorithmic techniques
to understand the genetic basis of diseases by identification of informative SNPs.
Our emphasis for SNP data analysis is combined approach on methods and techniques from Theoretical Computer Science, Analysis of Algorithms,
Software Engineering and Object-Oriented Programming, Experimental Algorithmics and Metaheuristics and their implementation in High-Performance
Distributed Computed Systems.
Metaheuristics for Life Sciences Problems
Project leaders:
C. Cotta and P. Moscato
Personnel involved:
R. Berretta, C. Cotta, A. Mendes and P. Moscato
Description: Metaheuristic methods are general purpose, high-level strategic techniques that are conceived and designed with the aim
of guiding other ad-hoc heuristics. Examples of metaheuristics include: Simulated Annealing, Tabu Search, and Memetic Algorithms. Moscato coined the latter
denomination in 1989 and has been working in the field since that year. Memetic Algorithms are now a well-recognized subfield of
Evolutionary Computation.
Within the aims of this subproject, special attention will be given to problems of interest to some other Life Science researchers
collaborating with the NBI in order to maximize cooperation and interdisciplinarity. The project will also develop new metaheuristics
for problems currently under investigation with researchers overseas (phylogenetic tree construction, feature subset selection,
data mining and clustering. etc). This project is complementary to other activities planned in the University of Newcastle
(i.e. the recently awarded ARC Centre of Excellence lead by Prof. John Aitken).
