Silico ADME Modeling

A simultaneous optimization of multiple parameters in carefully planned iterations is therefore required to arrive at molecules with suitable properties and profiles. 

The requirement for shortening the pharmaceutical discovery process caused the integration of pharmacokinetic and drug development efforts into earlier discovery to rapidly focus on molecules with increased optimisation potential. It is mandatory to initiate time-consuming optimisation programs only for those chemical series which show a potential to be convertible into drug molecules. 

The introduction of several new technologies like protein crystallography, ADME assays, medicinal 

In the following sections, this overview provides examples for both general and local models to illustrate different approaches for building models on structurally diverse or related training sets. Datasets for intestinal human absorption and human serum albumin binding will be discussed in detail while models for other relevant ADME properties have also been obtained. In addition, two local models will also be discussed addressing intestinal absorption from both an in vitro and an in vivo perspective. Those models, however, do not stand alone, but are used in combination with complementary models tailored for affinity and selectivity in the frame of multidimensional lead optimization. 

It has to be pointed out that prediction failures of general ADME models are often related to two major sources namely the quality of experimental data used to derive the model and the interpretation of the final model. These problems are discussed in depth by Stouch et al. (2003). Some models fail as they were built from data collected from different sources and laboratories. Although this might work for some robust standardized ADME assay, it could produce incomparable data for others. Such problems have been reported for example for Caco-2 assays from different laboratories. Even if the experimental 

---

Narayanan R and Gunturi SB. In silico ADME modelling prediction models for blood-brain barrier permeation using a systematic variable selection method. Bioorg Med Chem 2005 13 3017-28. 

In Silico ADME Modeling 3 Computational Models to Predict Human Intestinal Absorption Using Sphere Exclusion and kNN QSARMethods. QSAR Combinatorial Science, 5, 653-668. 

Gunturi, S.B., Narayanan, R. and Khandelwal, A. (2006) In silico ADME modelling 2 Computational models to predict human serum albumin binding affinity using ant colony systems. Bioorganic and Medicinal Chemistry, 14, 4118 129. 

The resulting semi-quantitative model was used in conjunction with structure-based docking and scoring, 3D-QS AR based affinity and selectivity predictions and in silico ADME models to estimate membrane permeability, solubility, and other key properties for the optimization process in this series. Hence, in this as well as in other series, multiple models can be collectively applied for ranking and prioritizing synthesis candidates and focused virtual libraries during advanced stages of multidimensional compound optimization. 

Gunturi, S.B. and Narayanan, R. (2007) In silico ADME modeling 3 computational models to predict human intestinal absorption using sphere exclusion... 

Hop, C.E. et al., High throughput ADME screening Practical considerations, impact on the portfolio and enabler of in silico ADME models, Curr. Drug Metab., 9(9), 847, 2008. 

In silico ADME modeling Predictive QSAR models for human intestinal absorption and oral bioavailability... 

---