Predictive models, drug-likeness

Bergstrom, C.A.S., Wassvik, C.M., Norinder, U., Luthman, K. and Artursson, P. (2004) Global and local computational models for aqueous solubility Prediction of drug-like molecules. Journal of Chemical Information and Computer Sciences, 44, 1477-1488. 

Bergstrom, C. A., C. M. Wassvik, U. Norinder, K. Luthman, and P. Artursson. 2004. Global and local computational models for aqueous solubility prediction of drug-like molecifle hem. Inf. Comput. Sci 44 1477-1488. 

Most practical implementations of drug-likeness use a computational model which takes as input the molecular structure, together with various properties, and predicts whether the molecule is drug-like or not. Some of these models may be very simple, such as a series of substructural filters. Only those molecules which pass all of these filters are output, Such filters can be used to eliminate molecules that contain inappropriate functionality. 

Another assessment of drug-likeness is afforded by Jorgensen s Qikprop family of ADME models [36]. In addition to the individual predictions, Jorgensen has proposed a rule-of-three. This proposes that a successful drug will have predicted... 

Two issues present themselves when the question of PB-PK model validation is raised. The first issue is the accuracy with which the model predicts actual drug concentrations. The actual concentration-time data have most likely been used to estimate certain total parameters. Quantitative assessment, via goodness-of-fit tests, should be done to assess the accuracy of the model predictions. Too often, model acceptance is based on subjective evaluation of graphical comparisons of observed and predicted concentration values. 

Although the pH-partition hypothesis and the absorption potential concept are useful indicators of oral drug absorption, physiologically based quantitative approaches need to be developed to estimate the fraction of dose absorbed in humans. We can reasonably assume that a direct measure of tissue permeability, either in situ or in vitro, will be more likely to yield successful predictions of drug absorption. Amidon et al. [30] developed a simplified film model to correlate the extent of absorption with membrane permeability. Sinko et al. [31] extended this approach by including the effect of solubility and proposed a macroscopic mass balance approach. That approach was then further extended to include facili-... 

The aim of the present example was to investigate whether the assessment of an in silico model of metabolic stability from a training set of several hundred drugs or drug-like compounds in human CYP3A4 cDNA-expressed microsomal preparations, would offer a suitable approach to predict the metabolic stability of external compounds. 

Singh, S.B., Shen, L.Q., Walker, M.J. and Sheridan, R.P. (2003) A model for predicting likely sites of CYP3A4-mediated metabolism on drug-like molecules. Journal of Medicinal Chemistry, 46, 1330-1336. 

The methodology is very fast and completely automated. To predict the site of metabolism for drug-like substrates, the method requires few seconds per molecule. It is important to point out that the method uses neither any training set nor any statistical model or supervised technique, and it has proven to be predictive for extensively diverse validation sets preformed in different pharmaceutical companies. 

A number of compounds with the lowest log BB values were overestimated by model 2. These entities were all drug-like compounds with a variety of structures and functional groups. Compounds at the other end of the scale, that is, compounds that exhibit high log BB values, are somewhat better predicted. These latter compounds were all hydrocarbons, for example, butanes, pentanes,... 

In the following text, a brief review of some recently published HIA models is presented. Recently, Hou and coworkers reported a quite large data set for HIA, which included 647 drug and drug-like molecules collected from a variety of literature sources. Among these 647 molecules, 578 are believed to be transported by passive diffusion. Based on this data set, Hou et al. [49,50] developed a set of prediction models for %HIA. 

Kortagere, S., Chekmarev, D., Welsh, W.J., Ekins, S. New predictive models for blood-brain barrier permeability of drug-like molecules. Pharm. Res. 2008, 25, 1836—45. 

---