A review on data mining and continuous optimization applications in computational biology and medicine.

摘要: An emerging research area in computational biology and biotechnology is devoted to mathematical modeling prediction of gene-expression patterns; it nowadays requests mathematics deeply understand its foundations. This article surveys data mining machine learning methods for an analysis complex systems biology. It mathematically deepens recent advances by rigorously introducing the environment aspects errors uncertainty into genetic context within framework matrix interval arithmetics. Given from DNA microarray experiments environmental measurements, we extract nonlinear ordinary differential equations which contain parameters that are be determined. done a generalized Chebychev approximation semi-infinite optimization. Then, time-discretized dynamical studied. By combinatorial algorithm constructs follows polyhedra sequences, region parametric stability detected. In addition, analyze topological landscape gene-environment networks terms structural stability. As second strategy, will review model selection kernel binary classification can used classify cancerous cells or discrimination other kind diseases. practically motivated theoretically elaborated; contribution better health care, progress medicine, education, more healthy living conditions.