Physicochemical descriptors estimation systems

Jarvis-Patrick clustering has also been used to support QSAR analysis in a system developed at the European Communities Joint Research Center. " ° The EINECS (European Inventory of Existing Chemical Substances) database contains more than 100,000 compounds and has been clustered using 2D structural descriptors. That database also has associated physicochemical properties and activities, but the data is very sparse. Jarvis-Patrick clustering was used to extract clusters containing sufficient compounds with measured data for an attempt to be made to estimate the properties of members of the cluster lacking the data. For a few clusters, it was used to develop reasonable QSAR models. 

Another approach for the mathematical modeling of the retention behaviour tries to estimate the retention times of a compound from experimental values obtained in the same system but under other chromatographic conditions. Estimation of the retention for a given compound should be easier and more accurate if we use its chromatographic data as a starting point, instead of structural descriptors or physicochemical parameters. 

One of the valuable options for development and application of the computational approach that is devoted to estimate endpoints related to nanomaterials is the employment of the above mentioned optimal descriptor which can represent a translator of eclectic information into the endpoint prediction. As an initial form of the optimal descriptor one can apply a hybrid of topological indices calculated with the adjacency matrix of the molecular graph and additive scheme. In this form a physicochemical parameter is modeled as the summation of contributions of molecular fragments. Further, the optimal descriptor could be expanded by means of taking into account contributions of various physicochemical conditions (presence/absence of defined chemical elements and/or defined types of covalent bonds, as well as different types of rings in the molecular system) which are able to modify the physicochemical (biochemical) behavior of a substance. The next step of evolution of the optimal descriptor is the involvement of the above mentioned eclectic factors related to various modes of action of the nanomaterials (Fig. 12.8). 

As discussed in Sect. 18.3, several models are available and can be applied with limited amounts of experimental data in combination with calculated molecular descriptors. These models are extremely valuable at the early stage of development of the API. A quick decision can be made regarding if ASD is an applicable approach for the candidate. Once this question is answered, the selection of an appropriate polymer comes in as the next step. If other additives are needed, for example, a surfactant to improve the manufacturability or to enhance the pharmacokinetic (PK) performance, the implication on system stability has to be assessed. The different formulations developed can be rank ordered based on their stability under severe temperature and humidity conditions. When the composition of the ASD is finalized, a more systematic stress test is used to understand the recrystallization risk. Though the changes in physicochemical properties by and large appear in a nonlinear fashion, the stress tests normally are sufficient to assess the risks, thus providing the formulation scientists confidence to estimate the stability of the drug product. 

---