Prospective Virtual Screening

Ballester, P. J., Westwood, I., Laurieri, N., Sim, E., Richards, W. G. (2010) Prospective virtual screening with Ultrafast shape recognition the identification of novel inhibitors of arylamine N-acetyltransferases. J R Soc Interface 7, 335-342. 

Perez-Nueno VI, Pettersson S, Ritchie DW et al (2009) Discovery of novel HIV entry inhibitors for the CXCR4 receptor by prospective virtual screening. J Chem Inf Model 49 810-823... 

Although complex cross-validation schemes are applied, the final assessment of the model quality and predictivity can only be achieved when the model is applied to compounds that were not seen by the model before. This so-called model validation is often the direct application of the model to prospective virtual screening projects or to predicting ADMET properties and the model quality may only be assessed after biological testing of the predicted compounds. 

The outcome of virtual screening is very much method, target, and data set dependent. It would be very helpful for practitioners to be able to predict the chance for success of a prospective virtual screening. Building models with machine... 

Retrospective Validation Prior to Prospective Virtual Screening... 

This chapter summarizes selected prospective virtual screening applications from the medicinal chemistry literature of the last decade. It does not attempt to provide... 

Although an increasing number of prospective virtual screening outcomes are now reported, most studies are still retrospective with a focus on novel software or virtual... 

Only prospective virtual screenings are shown purely retrospective studies or enrichment studies, as well as studies without experimental testing, are not included. The targets are grouped in different protein classes and sorted alphabetically within each class (first column). Where possible, the number of tested compounds, the number of detected active compounds, and activity data of the most potent hits are reported in the third column (n.a., not available). In general, the results refer to the primary screening hits secondary screens and optimizations are not considered. 

The other important aspect is the reliability of retrospective validations. Careful validations are critical to ascertain whether a proposed virtual screening technique will perform well when applied prospectively. This requires benchmarks that mimic key features of prospective virtual screens, including database size and the number of putative hits that can be tested in vitro. Unfortunately, our ability to generate retrospective tests that accurately capture the difficulty of prospective applications of virtual screening methods is still very limited [20]. Furthermore, the virtual lack of blind tests has contributed to the overfitting of current retrospective benchmarks by virtual screening techniques and thus to an overestimation of their prospective performance [25]. Regrettably, retrospective validations that do not address the... 

---