Quality of Fit and Predictive Ability of a continuous QSAR Model
According to A. Tropsha et al. (QSAR Comb. Sci. 22 (2003) 69-77 and Mol. Inf. 2010, 29, 476-488) the following statistical criteria must be satisfied by a predictive model:
1. R^2 > 0.6
2. Rcvext^2 > 0.5
3. (R^2 - R0^2)/R^2 < 0.1
4. (R^2 - R'0^2)/R^2 < 0.1
5. abs(R0^2 - R'0^2) < 0.3
6. 0.85 ≤ k ≤ 1.15
7. 0.85 ≤ k' ≤ 1.15
where:
R^2 Correlation coefficient between the predicted and observed activities
Rcvext^2 External cross validation
R0^2 Coefficient of determination: predicted versus observed activities
R'0^2 Coefficient of determination: observed versus predicted activities
k = slope: predicted versus observed activities regression lines through the origin
k’ = slope: observed versus predicted activities regression lines through the origin
More details and examples can be found here: www.novamechanics.com/knime.php
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.
Α. Afantitis, G. Melagraki, H. Sarimveis, P.A. Koutentis, O. Igglessi – Markopoulou and G. Kollias 'A combined LS-SVM and MLR QSAR workflow for predicting the inhibition of CXCR3 receptor by quinazolinone analogues' Molecular Diversity (2010) 14:225–235
- Type: TableIn-Port nameValues for the dependent variable, predicted by the model (ypred).
- Type: TableIn-Port nameValues for the dependent variable for the test set (yexp).
- Type: TableIn-Port nameValues for the dependent variable for the training set (ytr).