You are here: Home » Development » API » API 1.2 - current version » Applicability domain background

Applicability domain background

Background on applicability domain

What is the Applicability Domain of a Model

First of all, a model is a mapping of a physical system to certain assumptions. An applicability domain, is it self a model able to tell whether another system
satisfies these assumptions. Have the assumptions been posed explicitly it is easy to tell if these are satisfied; but this is not the case with statistical-based models.
The domain of applicability is a subset of the input space of the model in which the model is considered to apply properly and give reliable predictions. The estimation
of an applicability domain requires knowledge of the training set and, sometimes, the trained model itself meaning that in some cases the AD cannot be provided
prior to model training. A starting point for studying various topics related to the AD can be found in [1].

There are two views emerging on applicability domain of (Q)SAR models in OpenTox :

The predictive model itself provides estimation of applicability domain


Applicability domain is estimated by a procedure , separate from the predictive model


  • Descriptor-based methods  (leverage, ranges , distances, probability density, convex hull)
  • Methods, based on structural similarity


A third view also exist, which is usually based on expert judgment of mechanism/mode of action.  At some cases it could be formalized by expert defined rules.

'Mechanistic' applicability domain

A reference for the mechanistic approach on the applicability domain for QSAR Models is given in [10]





  1. J. Jaworska, N. Nikolova-Jeliazkova and Tom Aldenberg, "QSAR Applicability Domain Estimation by Projection of the Training Set in Descriptor Space: A review", ATLA 33, 445-459, 2005.
  2. Ellison, C. M.; Enoch, S. J.; Cronin, M. T. D.; Madden, J. C.; Judson, P., A structural fragment based approach to define the applicability domain of knowledge based predictive toxicology expert systems. ATLA-Alternatives to Laboratory Animals 2009, 37, 533-547.
  3. Chemical Domain of QSAR Models from Atom-Centered Fragments, Ralph Kuhne, Ralf-Uwe Ebert and Gerrit Schuurmann, J. Chem. Inf. Model., 2009, 49 (12), pp 2660–2669
  4. Horvath Dragos, Marcou Gilles, and Varnek Alexandre, Predicting the Predictability: A Unified Approach to the Applicability Domain Problem of QSAR Models, J. Chem. Inf. Model. 2009, 49, 1762–1776
  5. T. Scior, J.L. Medina-Franco, Q.-T. Do, K. Martínez-Mayorga, J. A. Yunes Rojas and P. Bernard, How to Recognize and Workaround Pitfalls in QSAR Studies: A Critical Review, Current Medicinal Chemistry, 2009, 16, 4297-4313
  6. Doweyko AM., QSAR: dead or alive?, J Comput Aided Mol Des. 2008 Feb;22(2):81-9.
  7. Stephen R. Johnson, The Trouble with QSAR (or How I Learned To Stop Worrying and Embrace Fallacy),J. Chem. Inf. Model. 2008, 48, 25-26 Read this!
  8. Elton Zvinavashe, Albertinka J. Murk, and Ivonne M. C. M. Rietjens, Promises and Pitfalls of Quantitative Structure-Activity Relationship Approaches for Predicting Metabolism and Toxicity, Chem. Res. Toxicol. 2008, 21, 2229–2236
  9. Tetko IV, Bruneau P, Mewes HW, Rohrer DC, Poda GI., Can we estimate the accuracy of ADME-Tox predictions? Drug Discov Today. 2006 Aug;11(15-16):700-7.
  10. Current status of methods for defining the applicability domain of (quantitative) structure-activity relationships. The report and recommendations of ECVAM Workshop 52., Netzeva TI, Worth A, Aldenberg T, Benigni R, Cronin MT, Gramatica P, Jaworska JS, Kahn S, Klopman G, Marchant CA, Myatt G, Nikolova-Jeliazkova N, Patlewicz GY, Perkins R, Roberts D, Schultz T, Stanton DW, van de Sandt JJ, Tong W, Veith G, Yang C., Altern Lab Anim. 2005 Apr;33(2):155-73.
  11. Guha R, Jurs PC., Determining the validity of a QSAR model--a classification approach,J. Chem. Inf. Model., 2005, 45 (1), pp 65–73. 
  12. Sabcho Dimitrov,Gergana Dimitrova,Todor Pavlov Nadezhda Dimitrova, Grace Patlewicz, Jay Niemela, and Ovanes Mekenyan, A Stepwise Approach for Defining the Applicability Domain of SAR and QSAR Models, J. Chem. Inf. Model., 2005, 45 (4), pp 839–849 
  13. D.W. Roberts et al, Mechanistic applicability domain classification of a local lymph node assay for skin sensitization, Chem Res Toxicol 2007, 20(7), 1019-30, PMID: 17555332



Document Actions