ALOGPS method

Thermodynamic descriptors AlogP98, logarithm of the octanol/water partition coefficient, atom-type value the AlogP98 descriptor is an implementation of the atom-type-based AlogP method using the latest published set of parameters. 

A set of 120 atom types was proposed. The corresponding hydrophobic constants were evaluated by multivariate regression analysis using a training set of 893 compounds, = 0.856 and s = 0.496. This approach is actually called the ALOGP method [Viswanadhan et al, 1993], and a similar approach was also proposed for - molar refractivity. As in the Broto-Moreau-Vandycke approach, correction factors are avoided, while hydrogen atoms are considered instead. 

K, V N Viswanadhan and J J Wendoloski 1998. Prediction of Hydrophobic (Lipophilic) lerties of Small Organic Molecules Using Fragmental Methods An Analysis of ALOGP and GP Methods. Journal of Physical Chemistry 102 3762-3772. 

Chose, A. K., Viswanadhan, V. N., Wendoloski, 1.1. Prediction of hydrophobic (lipophilic) properties of small organic molecules using fragmental methods an analysis of ALOGP and CLOGP Methods./. Phys. Chem. 1998, A 102, 3752-3772. 

Other difficulties arise from the incorrect treatment of aromaticity, e.g. the tautomeric form C can be also represented as an aromatic compound. However, ALOGPS, for example, does not consider this ring as an aromatic one. Inconsistently defined aromaticity lowers the prediction performance (Fig. 15.ID). The use of SDF files, which do not explicitly define aromaticity solves this problem. All these factors are not limiting when the data are prepared with the same coding scheme and consistency. However, these issues are very important for method application and benchmarking. 

However, there is still a strong need to develop new methods that will be able to quantitatively or at least qualitatively estimate the prediction accuracy of log D models. Such models will allow the computational chemist to distinguish reliable versus nonreliable predictions and to decide whether the available model is sufficiently accurate or whether experimental measurements should be provided. For example, when applying ALOGPS in the LIB RARY model it was possible to predict more than 50% and 30% compounds with an accuracy of MAE <0.35 for Pfizer and AstraZeneca collections, respectively [117]. This precision approximately corresponds to the experimental accuracy, s=0.4, of potentiometric lipophilicity determinations [15], Thus, depending on the required precision, one could skip experimental measurements for some of the accurately predicted compounds. 

Several methods considered in this article, such as the quantum-chemical approaches developed by Clark et al. [51,113,114] or the ALOGPS program [99], provide an estimation of the accuracy of model predictions. Other approaches to estimate accuracy of model predictions were reviewed elsewhere [99,141],... 

---