This project is designed to provide novel mathematical models and advanced
non-numerical algorithms for biological data analysis. A major goal is to
devise powerful new pattern recognition and molecular diagnostic software
for colorectal, breast, prostate, melanoma, lung and other high-incidence
cancers in Australia. Specific aims include
(1) the use of combinatorial optimization modeling to help understand
the foundational complexity of underlying problems in both the classical and
the parameterized settings,
(2) the application of state-of-the-art methods for the synthesis of
pre-processing rules for efficient extremal structures to aid genomic data
analysis,
(3) the development of hybrid algorithmic solutions based on a
combination of fixed-parameter tractable algorithms, integer programming and
evolutionary computation methods, and
(4) the study of how these innovative solutions can help answer questions
in genetics related to the analysis of gene expression and single nucleotide
polymorphisms in cancer research data.
