Construction of QSAR Models

Three sets of traditional models for toxicity using regression and discriminate analysis are generally produced. General models are often produced that rely on chemical parameters such as Log P. Models are also often produced that attempt to describe a particular subset of compounds unique in their composition or mode of action. Third, models can be produced that incorporate toxicity data from other species or other types of biological measurements. 

The first groups of models are generally constructed using molecular connectivity indices, kappa environmental descriptors, electronic charges, and substructural keys. In many instances Log P has been used however, our 

Daphnia EC50 Equation for Model Incorporating Molecular Connectivity Indices and Substructural Keys 

Primary amine bound to aromatic ring atom 1.0167 

Primay amine bound to aliphatic or alicycle carbon 1.0343 

---

The challenge here is experimental design. Scientists from the more experimental disciplines will probably already be aware of the need for design when planning an experiment, but it may not be obvious that the successful construction of QSAR models also calls for experimental design. Each compound tested or included in a QSAR analysis corresponds to a design point the experimental... 

Topological parameters were the first molecular descriptors that could be calculated for any chemical structure (other than obvious parameters such as molecular weight, atom counts, etc.) all that was required for their computation was a standard two-dimensional (2-D) representation of a chemical structure. The best known topological parameters are the molecular connectivity indices first described by Randic and investigated extensively by Hall and Kier et al. Molecular connectivity descriptors have since been used by chemists in the construction of QSAR models for many types of biological properties, especially applications in the environmental area. The reason for such heavy application in environmental studies is because these data sets often contain diverse sets of compounds, and as just mentioned, topological descriptors can be computed for any structure. 

Hopfinger et al. [53, 54] have constructed 3D-QSAR models with the 4D-QSAR analysis formahsm. This formalism allows both conformational flexibility and freedom of alignment by ensemble averaging, i.e., the fourth dimension is the dimension of ensemble sampling. The 4D-QSAR analysis can be seen as the evolution of Molecular Shape Analysis [55, 56]. 

What is the challenge for in silico ADME-Tox modeling In our point of view, the challenge is how to build up an accurate, reliable, and efficient model with good applicability. To answer this question, we may review the procedure of constructing an ADME-Tox model. The basic procedure of QSAR model construction is shown in Figure 2. 

Figure 2 The basic procedure of constructing a QSAR model.

A standard assumption in QSAR studies is that the models describing the data are linear. It is from this standpoint that transformations are performed on the bioactivities to achieve linearity before construction of the models. The assumption of linearity is made for each case based on theoretical considerations or the examination of scatter plots of experimental values plotted against each predicted value where the relationship between the data points appears to be nonlinear. The transformation of the bioactivity data may be necessary if theoretical considerations specify that the relationship between the two variables... 

From the above methods of constructing QSAR models several important parameters can be realized. The composition of the Training and Test Sets play an important role in the ability of the model to predict the bioactivities of the known and novel compounds. The methods used to set up the molecules, specifically the partial charges, can have a large impact on the information extracted from the model. The type of QSAR model to construct (traditional, 3D, D) will dictate the type of information gathered from the model. 

As has been discussed, there are several key steps to keep in mind when performing a QSAR study. They may seem minor details or trivial, yet all are important for obtaining a usable final model. From the molecules chosen for the Training and Test Sets to the number of descriptors used to create the model, all aspects of how the model was created are valuable in assessing the worthiness of the model or determining where errors may have occurred in construction of the model. The following are questions to ask when performing a QSAR study ... 

There are a number of QSAR approaches useful for predicting receptor binding affinity. These range from simple rejection filters for drug-like chemical identification to more sophisticated QSAR models used in lead optimization. We have constructed various types of QSAR models for ER binding. 

Figure 5.1 The general flow-chart of the MFTA QSAR analysis involves the construction of a model and the prediction of activity.

Role of Different Theoretical Methods Towards Constructing Effective QSAR Models... 

Among the various reactivity descriptors, electrophilicity [256-259] (constructing useful QSAR models. 

MOE Chemical Computing Group Construction of QSAR/QSPR models using probabilistic methods and decision trees, PCR and PLS methods http //www.chemcomp.com/software-chem.htm... 

The aim of this article is to provide an overview of the range of methodologies that can be used to build qualitative and quantitative models of DNA or RNA-ligand interactions. The construction of such models may resort to computational methods that go beyond 3D-SAR and 3E>-QSAR, and recent reviews of these techniques are available (see Free Energy Calculations Methods and Applications Free Energy Per-turbation Calculations Free Energy Simulations and Solvation Modeling). ... 

---