A QSAR

The derivation of a QSAR equation involves a number of distinct stages. First, it is obviousl necessary to synthesise the compormds and determine their biological activities. Whe planning which compormds to synthesise, it is important to cover the range of propertie that may affect the activity. This means applying the data-checking and -manipulation prc cedures discussed earlier. For example, it would be unwise to make a series of coinpound with almost identical partition coefficients if this is believed to be an important property. 

In order to parameterize a QSAR equation, a quantihed activity for a set of compounds must be known. These are called lead compounds, at least in the pharmaceutical industry. Typically, test results are available for only a small number of compounds. Because of this, it can be difficult to choose a number of descriptors that will give useful results without htting to anomalies in the test set. Three to hve lead compounds per descriptor in the QSAR equation are normally considered an adequate number. If two descriptors are nearly col-linear with one another, then one should be omitted even though it may have a large correlation coefficient. 

AJ Hopfinger. A QSAR investigation of dihydrofolate reductase inhibition by Baker triazmes based upon molecular shape analysis. I Am Chem Soc 102 7196-7206, 1980. 

Bis(oxazohnes) figands have been so widely used for the Diels-Alder reaction between N-2-alkenoyl-l,3-oxazolidine-2-one and cyclopentadiene that Lipkowitz and Pradhan developed a QSAR (quantitative structure-activity relationship) using Comparative Molecular Field Analysis (CoMFA) for a set of 23 copper-catalysts containing mainly bis(oxazoline) figands. The generated... 

Another important change started in the mid-1990s. Traditionally, a QSAR determined at a pharmaceutical company might have involved only 5-30 compounds. The size depended on how many compounds the medicinal chemist had synthesized and submitted to testing by the biologists. Sometimes this size data set sufficed to reveal useful trends. Other times, though, the... 

The information in the structures and known activity data is good enough to create a QSAR model with a confidence of 75% ... 

Kovatdieva A, Golbraikh A, Oloff S, Feng J, Zheng W, Tropsha A. QSAR modeling of datasets with enantioselective compounds using chirality sensitive molecular descriptors. SAR QSAR Environ Re. 2005 16(l-2) 93-102. 

In 1868 two Scottish scientists, Crum Brown and Fraser [4] recognized that a relation exists between the physiological action of a substance and its chemical composition and constitution. That recognition was in effect the birth of the science that has come to be known as quantitative structure-activity relationship (QSAR) studies a QSAR is a mathematical equation that relates a biological or other property to structural and/or physicochemical properties of a series of (usually) related compounds. Shortly afterwards, Richardson [5] showed that the narcotic effect of primary aliphatic alcohols varied with their molecular weight, and in 1893 Richet [6] observed that the toxicities of a variety of simple polar chemicals such as alcohols, ethers, and ketones were inversely correlated with their aqueous solubilities. Probably the best known of the very early work in the field was that of Overton [7] and Meyer [8], who found that the narcotic effect of simple chemicals increased with their oil-water partition coefficient and postulated that this reflected the partitioning of a chemical between the aqueous exobiophase and a lipophilic receptor. This, as it turned out, was most prescient, for about 70% of published QSARs contain a term relating to partition coefficient [9]. 

Despite the work of Overton and Meyer, it was to be many years before structure-activity relationships were explored further. In 1939 Ferguson [10] postulated that the toxic dose of a chemical is a constant fraction of its aqueous solubility hence toxicity should increase as aqueous solubility decreases. Because aqueous solubility and oil-water partition coefficient are inversely related, it follows that toxicity should increase with partition coefficient. Although this has been found to be true up to a point, it does not continue ad infinitum. Toxicity (and indeed, any biological response) generally increases initially with partition coefficient, but then tends to fall again. This can be explained simply as a reluctance of very hydrophobic chemicals to leave a lipid phase and enter the next aqueous biophase [11]. An example of this is shown by a QSAR that models toxicity of barbiturates to the mouse [12] ... 

Clearly, there are three basic steps to developing a QSAR ... 

Concerning the last point, Topliss and Costello [42] proposed that, to minimize the risk of chance correlations, a QSAR developed with MLR should utilize at least five data points (compounds) for each descriptor included in the equation. Later work [17] showed that it was necessary to take into account not only the number of descriptors in the QSAR (usually several) but also the whole of the descriptor pool (often several hundred) from which the best descriptors were selected. 

A QSAR for which the standard error of each descriptor is given concerns the bradycardic effect of a series of tetraalkylbispidines [47]. The QSAR models the selectivity between the desired bradycardic effect and the adverse contractile effect. It is important, in assessing and modeling drug toxicity, that the toxic effect is assessed relative to the desired effect as described above. The QSAR developed for the selectivity of the tetraalkylbispidines was ... 

A QSAR for the acute toxicity of new hypoglycemic agents [48] was internally cross-validated, but used LD50 instead of log LD50 as the dependent variable, and (more seriously) used LD50 values in g kg rather than in a molar unit such as mmol kg. ... 

A key requirement of QSAR is that the compounds used in the modeling and prediction processes should have the same mechanism of action, and for this reason most QSAR studies are made with congeneric series of compounds. However, if a diverse set of compounds can reasonably be assumed to have the same mechanism of action, QSAR modeling can justihably be carried out. For example, Dearden et al. [43] developed a QSAR for the ratio of brain levels of 22 very diverse drugs in the wild-type mouse and the P-glycoprotein knockout mouse (R+/ ) ... 

QSAR model validation is an essential task in developing a statistically vahd and predictive model, because the real utility of a QSAR model is in its ability to predict accurately the modeled property for new compounds. The following approaches have been used for the vahdation of QSAR Eqs. 1-20 ... 

BeUo-Ramirez AM, Nava-Ocampo AA (2004) A QSAR analysis of toxicity of Aconitinn alkaloids. Fund Clin Pharmacol 18 699-704... 

PSA is also used as H-bond descriptor to predict various properties of chemicals and drugs. PSA is defined as that part of a molecular surface that arises from oxygen and nitrogen atoms, and also the hydrogens attached to them. Applications of PSA as a QSAR descriptor in correlations with permeability and absorption were carried out first by Van de Waterbeemd et al. [30] and Palm et al. [31]. Clark [32-34] developed this further. Chapter 5 in this book is completely devoted to... 

Raevsky, O. A. QSAR description of molecular strucmre. In QSAR in Drug Design and Toxicology, Hadzi, D., Jerman-Blazic, B. (eds.), Elsevier, Amsterdam, 1987, pp. 31-36. 

It is usual to have the coefficient of determination, r, and the standard deviation or RMSE, reported for such QSPR models, where the latter two are essentially identical. The value indicates how well the model fits the data. Given an r value close to 1, most of the variahon in the original data is accounted for. However, even an of 1 provides no indication of the predictive properties of the model. Therefore, leave-one-out tests of the predictivity are often reported with a QSAR, where sequentially all but one descriptor are used to generate a model and the remaining one is predicted. The analogous statistical measures resulting from such leave-one-out cross-validation often are denoted as and SpR ss- Nevertheless, care must be taken even with respect to such predictivity measures, because they can be considerably misleading if clusters of similar compounds are in the dataset. 

However, Clog P and, more generally, Hpophilidty descriptors referring to octanol-water are not the only lipophilicity parameters to be taken into account As mentioned above, isotropic and anisotropic Hpophilidty values gave rise to two different Hpophilidty scales for ionized compounds and thus it is recommended to test both of them (after checking the absence of any coHnearity) when looking for a QSAR model involving ions. 

The basic hypothesis of a QSAR model is that the activity (or effect or property) can be put in relationship with the chemical, using some parameters to describe the chemical. Thus, the three main components of the QSAR model are the activity to be modeled, the chemical information, and the way to establish a link between these two components. For this, we need some suitable ways to describe the chemical and a good mathematical algorithm. 

Typically a QSAR model is built up starting with a set of chemicals with known property values. 

Let s discuss the first requirement, a criterion for us. We notice that it is not requested that the model is validated. Validation is a formal process, which takes many years. The formal validation process of a QSAR model would end after REACH probably. 

OECD also provides a check list for the application of its principles in the context of QSAR validation. This checklist can be useful to help scientists and regulators during the selection of a QSAR model and to evaluate its robustness/ validity [9]. 

Finally, a QSAR evaluation of different chemicals from waste-related products and recycling is shown in order to underline how in silico models can be used as a valid tool to fill in the gaps and to obtain information on toxicological profile and physicochemical information on compounds. In particular, a focus on compounds suggested by EU project Riskcycle is presented. 

---