The International Journal of Biochemistry & Cell Biology
ReviewApplications of genetic programming in cancer research
Section snippets
Background
Darwin's fundamental insight into the nature of evolution has been one of the cornerstones of modern biology. The diversity of species and individuals in the fossil record shows the evolution of complex features, not once, but many times, demonstrating that evolution is a powerful, creative force. Starting in the late 1950s, computer scientists began using the concepts of evolution and natural selection to find creative solutions for computing problems (Fraser, 1957, Barricelli, 1957,
Cancer treatment and research
Genetic programming has been used in a number of areas of cancer research as well as clinical treatment. The advent of modern molecular techniques such as microarrays, proteomics and single nucleotide polymorphism arrays (SNP chips) has provided a large amount of interrelated data that is not easily modeled using conventional data analysis methods. Genetic programming has been used alone, and in combination with other methods to produce useful and informative analyses. Some of these
Other applications
In addition to the cancer-specific papers outlined above, GP has applications in other areas that may have an impact on cancer research and treatment in the future. This includes studies on drug discovery and development (Langdon, 2008a, Archetti et al., 2007) and modeling of dynamic processes such as deducing a metabolic pathway from empirical data (Koza et al., 2001). Reif et al. (2004) demonstrate GP's ability to integrate diverse types of complex molecular data to produce a model based in
Conclusion
Genetic programming has a growing record of success in the analysis of large, complex data sets associated with cancer both from a research perspective and from a clinical context. Its abilities to find useful combinations of features from molecular data set in an unbiased way and to produce human comprehensible non-linear models to describe disease states and characteristics make it particularly useful for such analyses.
Though the development of artificial populations competing within a
Disclosures
W.P.W serves as Genetics Squared's Chief Executive Officer and A.M.C is a member of Scientific Advisory Board for Genetics Squared, Inc.
References (47)
Genetic Algorithm Learning and the Cobweb Model
J Econ Dyn Control
(1994)- et al.
A constrained-syntax genetic programming system for discovering classification rules: application to medical data sets
Artif Intell Med
(2004) - et al.
MALDI-TOF mass spectrometry analysis of cerebrospinal fluid tryptic peptide profiles to diagnose leptomeningeal metastases in patients with breast cancer
Mol Cell Proteomics
(2005) - et al.
Genetic algorithm for large-scale maximum parsimony phylogenetic analysis of proteins
Biochim Biophys Acta
(2005) - et al.
Intercellular adhesion molecule-1 (ICAM-1) expression and cell signaling cascades
Free Radic Biol Med
(2000) - et al.
Program structure-fitness disconnect and its impact on evolution in GP
- et al.
Genetic programming for computational pharmacokinetics in drug discovery and development
Genet Program Evol Mach
(2007) Symbiogenetic evolution processes realized by artificial methods
Methods
(1957)- et al.
Stock selection: an innovative application of genetic programming methodology
Optimization through evolution and recombination