Applicability Domain (APD) based on the extent of extrapolation

Domain of model applicability must be defined to flag compounds in the test set for which predictions may be unreliable. In this node extent of extrapolation method is used to define the domain of applicability of the model that could be used for screening new compounds. Predictions for only those compounds that fall into this domain may be considered reliable.

Extent of extrapolation is one simple approach to define the applicability of the domain [A. Tropsha, P. Gramatica, V.K. Gombar, QSAR Comb. Sci. 22 (2003) 69-77 & A. Golbraikh, A. Tropsha, J. Mol. Graph. Model. 20 (2002) 269-276]. It is based on the calculation of the leverage hi for each chemical, where the in silico model is used to predict its activity property:

hi = xiT(XTX)xi

The descriptor-row vector of the query compound and X is the k x n matrix containing the k descriptor values for each one of the n training compounds. A leverage value greater than 3k/n is considered large. It means that the predicted response is the result of a substantial extrapolation of the model and may not be reliable.

If this node is useful to you, please cite the following papers:

G. Melagraki, Α. Afantitis, H. Sarimveis, P.A. Koutentis, O. Igglessi – Markopoulou and G. Kollias "Predictive QSAR workflow for the in silico identification and screening of novel HDAC inhibitors" Molecular Diversity 13 (3), pp. 301-311. (link)

Α. Afantitis, G. Melagraki, H. Sarimveis, P.A. Koutentis, J. Markopoulos and O. Igglessi – Markopoulou "Development and Evaluation of a QSPR Model or the Prediction of Diamagnetic Susceptibility" QSAR Comb. Sci. 27, 2008, No. 4, 432 – 436. (link)

KNIME Node Options:

Input Ports
0 Training Set (Please insert only selected attributes)
1 Test or Virtual Screening Set (Please insert the same attributes with Training Set)

Output Ports
0 Leverage for each compound of the Test/ Virtual Screening Set, Leverage Limit for the Training Set, Result: Reliable / Unreliable

Each Enalos Domain of Applicability node provides a summary View with information about the percentage of the reliable and the unreliable predictions.

Download Enalos Domain Leverage Node from:  KNIME Update Community Contributions Nightly

Instructions:  Open Knime, Go to Help, Install New Software, From "Available Software Sites" Select Stable Community Contributions, Select Cheminformatics (see figure) and then Select Enalos Nodes for KNIME (Enalos_nodes_for_KNIME.png)

Download Sample Workflow (Enalos Domain Leverages Node): (link)