Quantitative structure-activity model validation

The abbreviation QSAR stands for quantitative structure-activity relationships. QSPR means quantitative structure-property relationships. As the properties of an organic compound usually cannot be predicted directly from its molecular structure, an indirect approach Is used to overcome this problem. In the first step numerical descriptors encoding information about the molecular structure are calculated for a set of compounds. Secondly, statistical methods and artificial neural network models are used to predict the property or activity of interest, based on these descriptors or a suitable subset. A typical QSAR/QSPR study comprises the following steps structure entry or start from an existing structure database), descriptor calculation, descriptor selection, model building, model validation. 

Basak, S. C., Mills, D., Hawkins, D. M., Kraker, J. J. Quantitative structure-activity relationship (QSAR) modeling of human blood air partitioning with proper statistical methods and validation. Chem. Biodivers., accepted. 

The chemometric basic tools may be divided into the following typologies of study data exploration, modelling, prediction and validation, design of experiments (DOE), process analytical technology (PAT), quantitative structure-activity relationship (QSAR). Details and relevant literature are reported in the following paragraphs. 

OECD. 2007b. Guidance document on the validation of (quantitative) structure-activity relationships [(Q) SAR] models. ENV/JM/MONO(2007)2. 15 February 2007. Paris OECD http //www.oecd.org/ searchResult/0,2665,en 2649 201185 l l l l L00.html OECD. 2007c. Report on the regulatory uses and applications in OECD member countries of (quantitative) structure activity relationship (Q)SAR models in the assessment of new and existing chemicals. ENV/JM/MONO(2006)25. 15 February 2007. Paris OECD http //www.oecd.org/searchResult/ 0,2665,en 2649 201185 l l l l L00.html... 

The validity of a model is always limited to a certain domain in the parameter space. For example, if a quantitative structure-activity relationships (QSAR) model is specified for nonpolar organic chemicals in the log range from 2 to 6 and has a molecular weight below 700, then an application to substances outside this range is an improper extrapolation. Note that the parameter space may be difficult to discern for example, combinations of low values for one variable and high values for another could constitute an extrapolation if such combinations had been missing in the validation or specification of the model. Exceedence of model boundaries introduces additional uncertainty at best, but can also lead to completely incorrect outcomes. 

Waller CL McKinney JD (1995) Three-dimensional quantitative structure activity relationships of dioxins and dioxin-like compounds Model validation and Ah receptor characterization. Chem Res Toxicol, 8 847-858. 

Peterson, Y. K., Wang, X. S., Casey, P. J., Tropsha, A. (2009) Discovery of geranylgeranyltransferase-I inhibitors with novel scaffolds by the means of quantitative structure-activity relationship modeling, virtual screening, and experimental validation. J Med Chem 52, 4210-4220. 

There are several properties of a chemical that are related to exposure potential or overall reactivity for which structure-based predictive models are available. The relevant properties discussed here are bioaccumulation, oral, dermal, and inhalation bioavailability and reactivity. These prediction methods are based on a combination of in vitro assays and quantitative structure-activity relationships (QSARs) [3]. QSARs are simple, usually linear, mathematical models that use chemical structure descriptors to predict first-order physicochemical properties, such as water solubility. Other, similar models can then be constructed that use the first-order physicochemical properties to predict more complex properties, including those of interest here. Chemical descriptors are properties that can be calculated directly from a chemical structure graph and can include abstract quantities, such as connectivity indices, or more intuitive properties, such as dipole moment or total surface area. QSAR models are parameterized using training data from sets of chemicals for which both structure and chemical properties are known, and are validated against other (independent) sets of chemicals. 

E. Benfenati et al., Validation of the Models, in Quantitative Structure—Activity Relationships (QSAR) for Pesticide Regulatory Purposes, ed. by E. Benfenati (Elsevier, Amsterdam, 2007), pp. 185-199... 

OECD. OECD Principles for the Validation, for Regulatory Purposes, of (Quantitative) Structure-Activity Relationship Models. Paris, France, http //www.oecd.org/document/23/ 0,3343,en 2649 34379 33957015 l l l l,00.html... 

Guidance Document on the Validation of (Quantitative) Structure Activity Relationship [(Q) SAR] Models, No. 69 OECD Series on Testing and Assessment Organisation of Economic Cooperation and Development Paris, France, 2007. http //www.oecd.org. Accessed May 2008... 

Organisation for Economic Co-operation and Development [OECD]. 2005a. OECD principles for the validation, for regulatory purposes, of (quantitative) structure-activity relationship models. http //www.oecd.Org/document/23/0,2340,en 2649 34365 33957015 l l l l,00.html (accessed December 15, 2005). 

This book intends to provide a starting point for those interested in the prediction of the toxicity and fate of chemicals to humans and the environment. SARs and, more frequently, quantitative structure-activity relationships (QS ARs) provide methods to predict these endpoints. A brief history of the area, the driving forces, and basis of the topic is provided in this chapter. Further chapters (2 to 7) describe the methods to develop predictive models the application of models to human health endpoints (Chapters 8 to 11) their application to environmental toxicity and fate (Chapters 12 to 17) and the use of predictive models (Chapter 19), adoption by the regulatory authorities (Chapter 19), and validation (Chapter 20). 

Quantitative structure-activity/pharmacokinetic relationships (QSAR/ QSPKR) for a series of synthesized DHPs and pyridines as Pgp (type I (100) II (101)) inhibitors was generated by 3D molecular modelling using SYBYL and KowWin programs. A multivariate statistical technique, partial least square (PLS) regression, was applied to derive a QSAR model for Pgp inhibition and QSPKR models. Cross-validation using the leave-one-out method was performed to evaluate the predictive performance of models. For Pgp reversal, the model obtained by PLS could account for most of the variation in Pgp inhibition (R2 = 0.76) with fair predictive performance (Q2 = 0.62). Nine structurally related 1,4-DHPs drugs were used for QSPKR analysis. The models could explain the majority of the variation in clearance (R2 = 0.90), and cross-validation confirmed the prediction ability (Q2 = 0.69) [ 129]. 

Chemical Classes Included in the RSTS The chemical and physical forms of the substances included must be consistent with the stated prediction model. For example, if the prediction model indicates that the alternative method is valid for assessing the eye irritation potential of mild, moderate, and severely irritating liquid, surfactant-based formulations, then the RSTS should contain liquid surfactant-based substances of the relevant class that cover a range of toxicity from mild to severe. Quantitative structure-activity relationships may be useful in helping selection of relevant test chemicals. 

In studies of quantitative structure activity relationships (QSAR), the relative potencies of a series of drugs are subjected to analysis with the hope that biological potency will be described by a mathematical equation. QSAR is an actuarial or statistical method in which only objective data are used with no intrusion of models or mechanistic hypotheses. The equation that is obtained not only accounts for the relative potencies of the compounds, but from it are deduced predictions of the potencies of untested compounds if the equation is valid, the predictions are ineluctable. The method thus has the capacity of yielding new (structurally related) drugs with desired potency, perhaps drugs with enhanced selectivity or fewer side effects. 

Kraker JJ, Hawkins DM, Basak SC, Natarajan R, Mills D. Quantitative structure-activity relationship (QSAR) modeling of juvenile hormone activity Comparison of validation procedures. Chemom Intel Lab Sys, in press, 2006. 

I. Motoc, R. A. Dammkoehler, D. Mayer, andj. Labanowski, Quant. Structure-Activity Relation. 5, 99 (1986). Three-Dimensional Quantitative Structure-Activity Relationships, 1. General Approach to the Pharmocophore Model Validation. 

All developments of quantitative structure activity relationships (QSARs)/ quantitative structure-property relationships (QSPRs)/QSDRs go through similar steps (1) collection of a database of measured values for model development and validation/evaluation, (2) selection of chemical descriptors (can include connection indices, atom, bond, or functional groups, molecular orbital calculations), (3) development of the model (develop a correlation between the chemical descriptors and the activity/property/degradation values) using a variety of statistical approaches (linear and non-linear regression, neural networks, partial least squares (PLS), etc. [9]), and (4) validate/evaluate the model for predictability (usually try to use a separate set of chemicals other than the ones used to train the model external validation) [10]. 

OECD (2007) OECD (Organization for Economic Co-operation and Development) Guidance document [ENV/TM/MONO(2007)2] on the validation of (Quantitative) structure-activity relationship [(Q)SAR] models. OECD Environment Health and Safety Publications (2007) Series on Testing and Assessment, No. 69, Paris... 

In summary, the support vector machine (SVM) and partial least square (PLS) methods were used to develop quantitative structure activity relationship (QSAR) models to predict the inhibitory activity of nonpeptide HIV-1 protease inhibitors. Cenetic algorithm (CA) was employed to select variables that lead to the best-fitted models. A comparison between the obtained results using SVM with those of PLS revealed that the SVM model is much better than that of PLS. The root mean square errors of the training set and the test set for SVM model were calculated to be 0.2027, 0.2751, and the coefficients of determination (R2) are 0.9800, 0.9355 respectively. Furthermore, the obtained statistical parameter of leave-one-out cross-validation test (Q ) on SVM model was 0.9672, which proves the reliability of this model. Omar Deeb is thankful for Al-Quds University for financial support. 

---