Selection Of Models For The Regression Problems In Biology Using Chi-Square Test

Abstract

Biological systems, at all scales of organization from nucleic acids to ecosystems, are inherently complex and variable. Therefore mathematical models have become an essential tool in systems biology, linking the behavior of a system to the interaction between its components. Parameters in empirical mathematical models for biology must be determined using experimental data, a process called regression because the experimental data are noisy and incomplete. The term "regression" dates back to Galton's studies in the 1890s. Considering all this, biologists, therefore, use statistical analysis to detect signals from the system noise. Statistical analysis is at the core of most modern biology and many biological hypotheses, even deceptively. Regression analysis is used to demonstrate association among the variables believed to be biologically related and fit the model to give the best model. There are two types of regression, linear and nonlinear regression to determine the best fit of the model. In this manuscript, we perform a least squares error fit to different models and select the best fit model using the chi-test, and determine the p-value of the selected model to data that was collected when various doses of a drug were injected into three animals, and the change in blood pressure for each animal was recorded.