The Applicability Domain toolbox is a collection of MATLAB modules for calculating the Applicability Domain (AD) of a model for a set of test samples and verify if they fall inside or outside the model AD. For a test sample outside the AD, the model shouldn’t be applicable since the prediction reliability will be low.
The toolbox implements a set of AD approaches based on several strategies, such as: bounding box, bounding box on PCs, convex hull, leverage, distance to centroid, k Nearest Neighbours (kNN) approach with fixed k, k Nearest Neighbours (kNN) approach with variable k, Probability Density Function-based methods. A graphical user interface (GUI), which allows an easy AD evaluation and analysis of results, is also provided.
Conditions and warranty
The toolbox is freeware and may be used (but not modified) if proper reference is given to the authors. Preferably refer to the followign papers:
F. Sahigara, D. Ballabio, R. Todeschini, V. Consonni (2013). Assessing the validity of QSARs for ready biodegradability of chemicals: An Applicability Domain perspective. Current Computer-Aided Drug Design, 10, 137-147
F. Sahigara, K. Mansouri, D. Ballabio, A. Mauri, V. Consonni, R. Todeschini, (2012) Comparison of different approaches to define the Applicability Domain of QSAR models, Molecules, 17, 4791-4810
In short, no guarantees, whatsoever, are given for the quality of this toolbox or for the consequences of its use. It is inevitable that there will be some bugs, but we have tried to test the algorithms thoroughly.
Download
Fill in the following form. Your personal data will be used only for notification via email of new releases of the toolbox and will not be communicated to external third parties. Once the form has been submitted, download the rar file containing the toolbox on a local folder, unzip the rar file and extract all the Matlab modules in a unique folder. Before starting calculations, please read the HELP files provided in HTML format. A complete guide on how to calculate models is provided.