Evolutionary methods for variable selection in the epidemiological modeling of cardiovascular diseases.

BACKGROUND: The redundancy of information is becoming a critical issue for epidemiologists. High-dimensional datasets require new effective variable selection methods to be developed. This study implements an advanced evolutionary variable selection method which is applied for cardiovascular predictive modeling. The epidemiological follow-up study KIHD (Kuopio Ischemic Heart Disease Risk Factor Study) was used to compare the designed variable selection method based on an evolutionary search with conventional stepwise selection. The sample contains in total 433 predictor variables and a response variable indicating incidents of cardiovascular diseases for 1465 study subjects. RESULTS: The effectiveness of variable selection methods was investigated in combination with two models: Generalized Linear Logistic Regression and Support Vector Machine. We managed to decrease the number of variables from 433 to 38 and save the predictive ability of the models used. Their performance was evaluated with an F-score metric. At most, we gained 65.6% and 67.4% of the F-score before and after variable selection respectively. All the results were averaged over 5-folds of a cross-validation procedure. CONCLUSIONS: The presented evolutionary variable selection method allows a reduced set of variables to be chosen which are relevant to predicting cardiovascular diseases. A reference list of the most meaningful variables is introduced to be used as a basis for new epidemiological studies. In general, the multicollinearity of variables enables different combinations of predictors to be used and the same performance of models to be attained.

The evolutionary origin of form and function.

We regard the basic unit of the organism, the cell, as a complex dissipative natural process functioning under the second law of thermodynamics and the principle of least action. Organisms are conglomerates of information bearing cells that optimise the efficiency of energy (nutrient) extraction from its ecosystem. Dissipative processes, such as peptide folding and protein interaction, yield phenotypic information from which form and function emerge from cell to cell interactions within the organism. Organisms, in Darwin's 'proportional numbers', in turn interact to minimise the free energy of their ecosystems. Genetic variation plays no role in this holistic conceptualisation of the life process.

Epigenetic regulation of the mammalian cell.

BACKGROUND: Understanding how mammalian cells are regulated epigenetically to express phenotype is a priority. The cellular phenotypic transition, induced by ionising radiation, from a normal cell to the genomic instability phenotype, where the ability to replicate the genotype accurately is compromised, illustrates important features of epigenetic regulation. Based on this phenomenon and earlier work we propose a model to describe the mammalian cell as a self assembled open system operating in an environment that includes its genotype, neighbouring cells and beyond. Phenotype is represented by high dimensional attractors, evolutionarily conditioned for stability and robustness and contingent on rules of engagement between gene products encoded in the genetic network. METHODOLOGY/FINDINGS: We describe how this system functions and note the indeterminacy and fluidity of its internal workings which place it in the logical reasoning framework of predicative logic. We find that the hypothesis is supported by evidence from cell and molecular biology. CONCLUSIONS: Epigenetic regulation and memory are fundamentally physical, as opposed to chemical, processes and the transition to genomic instability is an important feature of mammalian cells with probable fundamental relevance to speciation and carcinogenesis. A source of evolutionarily selectable variation, in terms of the rules of engagement between gene products, is seen as more likely to have greater prominence than genetic variation in an evolutionary context. As this epigenetic variation is based on attractor states phenotypic changes are not gradual; a phenotypic transition can involve the changed contribution of several gene products in a single step.

Modeling the C. elegans nematode and its environment using a particle system.

A particle system, as understood in computer science, is a novel technique for modeling living organisms in their environment. Such particle systems have traditionally been used for modeling the complex dynamics of fluids and gases. In the present study, a particle system was devised to model the movement and feeding behavior of the nematode Caenorhabditis elegans in three different virtual environments: gel, liquid, and soil. The results demonstrate that distinct movements of the nematode can be attributed to its mechanical interactions with the virtual environment. These results also revealed emergent properties associated with modeling organisms within environment-based systems.

An artificial ecosystem: emergent dynamics and lifelike properties.

We discuss modeling and analysis of an artificial ecosystem. The ecosystem consists of basic elements, scents, plants, and animals. There are two species of animals: worms and beetles. As beetles absorb energy from worms, which absorb energy from blades of grass, which absorb energy from water, there is a food chain connecting animals to basic elements. The novelty of our approach lies in the modeling technique: we model the entire ecosystem using a single particle system. Consequently, the physical interaction dynamics not only shows emergent dynamics, but also some interesting lifelike properties. As the main contribution, we formalize the particle system and use it to model and analyze the ecosystem. We consider here several scenarios with nontrivial interaction dynamics.