Prediction of physicochemical properties

Pharmacokinetics and toxicity have been identified as important causes of costly late-stage failures in drug development. Hence, physicochemical as well as ADMET properties need to be fine-tuned even in the lead optimization phase. Recently developed in silica approaches will further increase model predictivity in this area to improve compound design and to focus on the most promising compounds only. A recent overview on ADME in silica models is given in Ref [128]. 

Currently available solubility models based on turbidimetry as well as nephelometry are not very predictive and are limited in their broad applicability, because they make use of training data from different laboratories determined under varying experimental conditions [130]. However, the access to many aqueous solubilities measured under standardized conditions is expected to greatly improve currently available models. 

---

Schtiurmann, G., Ebert, R. U Ktihne, R. Prediction of physicochemical properties of organic compounds from 2D molecular structure - fragment methods vs. LEER models. Chimia 2006, 60, 691-698. 

Morris, J. J., Bruneau, P. Prediction of physicochemical properties. In Virtual Screening for Bioactive Molecules, Bohm,... 

VII. Appendix 1. Commonly Available Methods for the Prediction of Physicochemical Properties SRC EPIWIN Software... 

VII. APPENDIX 1. COMMONLY AVAILABLE METHODS FOR THE PREDICTION OF PHYSICOCHEMICAL PROPERTIES SRC EPIWIN SOFTWARE... 

Structure-activity relationships (SARs) and quantitative structure-activity relationships (QSARs), referred to collectively as QSARs, can be used for the prediction of physicochemical properties, environmental fate parameters (e.g., accumulation and biodegradation), human health effects, and ecotoxicological effects. A SAR is a (qualitative) association between a chemical substructure and the potential of a chemical containing the substructure to exhibit a certain physical or biological effect. A QS AR is a mathematical model that relates a quantitative measure of chemical structure (e.g., a physicochemical property) to a physical property or to a biological effect (e.g., a toxicological endpoint). 

Furthermore, in-silico models need to be challenged with new data, since models derived from small training datasets may become unstable upon the addition of new compounds. Finally, the prediction of physicochemical properties should be checked against experimental data for representative members of a lead series. 

Jurs, PC, Hasan, M.N., Hansen, PJ. and Rohrbaugh, R.H. (1988). Prediction of Physicochemical Properties of Organic Compounds from Molecular Structure. In Physical Property Prediction in Organic Chemistry (Jochum, C., Hicks, M.G. and Sunkel, J., eds.). Springer-Verlag, Berlin (Germany), pp. 209-233. 

Taskinen J, Yliruusi J. Prediction of physicochemical properties based on neural network modelling. Adv Drug Deliv Rev 2003 55 1163-83. 

Artemenko NV, Baskin II, Palyulin VA, Zefirov NS. Artificial neural network and fragmental approach in prediction of physicochemical properties of organic compounds. Russ Chem Bull 2003 52 20-9. 

Morris JJ, Bruneau PP. Prediction of physicochemical properties. In Bohm HJ, Schneider G, editors, Virtual screening for bioactive molecules. Weinheim Wiley-VCH, 2000. p. 33-58. 

Quantum-mechanical studies on the tautomerism of heterocyclic compounds involve, in general, two aspects. The first deals with the prediction of physicochemical properties of defined tautomeric forms (e.g., ultraviolet spectra, dipole moments, ionization potentials, etc.). This seems to be easy to handle. Using any semiempirical or nonempirical quantum-mechanical computational method, depending on approximations involved in the method, we are able to calculate properties that, more or less, agree with experimental values. Calculations of this type do not contribute to a direct estimation of the relative stability of the tautomers, however they are particularly important for cases in which a tautomeric form of a compound is so rare that it is not possible to measure it directly. 

The molecular structure is at the basis of physicochemical, DMPK, and safety/ toxicity properties as outlined in Figure 5.1. Measurement and prediction of physicochemical properties are relatively easy compared to those of DMPK and safety properties, where biological factors come into play. However, DMPK and toxicity properties depend to a certain extent on the physicochemical properties of compounds as these dictate the degree of access to biological systems such as enzymes and transporters. 

Many of the physicochemical properties of interest are dependent on the solid form and, unfortunately, successful prediction of polymorphic forms is inexact. This, in combination with the fact that prediction of physicochemical properties is also very challenging, makes ah initio prediction very difficult and imprecise. However, some discussion of predictive tools is included in this chapter. A general comment regarding ah initio prediction is that "order of magnitude" predictions may be possible once some basic physicochemical information is available. However, the complexity and diversity of the chemistry space make reliable predictions across a broad spectrum of chemical structures very difficult. It is not surprising then that physicochemical predictions across more narrowly defined chemical spaces (e.g., chemical or therapeutic classes) can be more reliable and useful. Drug delivery, formulation, and computational chemistry experts will likely be able to provide a useful perspective on opportunities to take advantage of such ah initio predictions within the chemistry space that discovery teams often operate. 

Valko K. Measurements and predictions of physicochemical properties. High-Throughput ADMETox Estimation 2002 1-24 A1-A5. 

Schneider, Eds, Wiley-VCH, Chichester, United Kingdom, 2000, pp. 33-58. Prediction of Physicochemical Properties. 

---