Model predictive in silico

Aronov, A.M. (2005) Predictive in silico modeling for hERG channel blockers. Drug Discovery Today, 10, 149-155. 

Virtual hits need to be synthesized for hits from virtual libraries and their bioactivities experimentally verified for VS to have any real impact. More importantly, these virtual hit followup steps can act as a validation stage for the computational models and the associated VS protocol. The results of experimental verification can be fed back to the in silico assay stage for building better predictive in silico models. 

Recently predictive in silico modeling for hERG channel blockers has been described." Different approaches have aimed primarily at fdtering out potential hERG channel blockers in the context of combinatorial and virtual libraries and to elucidate structure-activity relationships. These new computational methods may predict trends, but are not as yet sufficiently precise to make valid predictions. 

This chapter will try to answer some of these questions and investigate various approaches to derive statistically sound, robust, and predictive in silico models. 

An integral part of deriving a statistically valid and predictive in silico model is the choice of training and test set, as well as the model validation. Without proper validation of the derived model, it is difficult to assess its statistical qualities and... 

The logical progression is in considering how best to improve a model s performance by considering what additional data to include. A study by Di Veroli et al. (2013) considers the kinetics of drug s block that can influence use and voltage dependence and how this may lead to the development of more predictive in silico models. It is... 

Compound optimization requires more sophisticated (quantitative) approaches. Large efforts have been undertaken to develop predictive in silico models, which support the progress of compound series in hit evaluation and in lead optimization. 

---