An efficient piecewise linear model for predicting activity of caspase-3 inhibitors

The linear (Multiple linear regression; MLR), non-linear (Artificial neural network; ANN), and an approach based on “Extended Classifier System in Function approximation” (XCSF) were applied herein to model the biological activity of 658 caspase-3 inhibitors.Various kinds of molecular descriptors were calculated to represent the molecular structures of the compounds. The original data set was partitioned into the training and test sets by the K-means classification method. Prediction error on the test data set indicated that the XCSF as a local model estimates caspase-3 inhibition activity, better than the global models such as MLR and ANN. The atom-centered fragment type CR2X2, electronegativity, polarizability, and atomic radius and also the lipophilicity of the molecule, were the main independent factors contributing to the caspase-3 inhibition activity.The results of this study may be exploited for further design of novel caspase-3 inhibitors.There are some challenges such as new disease and drug resistance which our society is faced with. On the other hand, drug discovery is a costly and time consuming process. In this context, there is a great demand for predictive models to design new drugs with improved properties and diminished side effects [1]. Furthermore, there is also a demand for new methods that replace and reduce the use of laboratory animals [2]. These methods should be used in the design and evaluation of experimental tests and in the selection of appropriate test compounds for validation studies [3].In order to reduce the time and cost during the drug development process, quantitative structure activity relationship (QSAR) is a robust scientific method which can help scientists to predict the activity and side effects of new compounds [4,5]. Distribution of training data affects on the model that is suitable for it, so searching to find a convenient method is an essential part of QSAR model evaluation to predict the biological activity for new tar