Validated QSAR Models as Virtual Screening Tools

Validated QSAR Models as Virtual Screening Tools 

Although combinatorial chemistry and HTS have offered medicinal chemists a much broader range of possibilities for lead discovery and optimization, the number of chemical compounds that can be reasonably synthesized, which is sometimes called virtual chemistry space , is stiU far beyond today s capability of chemical synthesis and biological assay. Therefore, medicinal chemists continue to face the same problem as before which compounds should be chosen for the next round of synthesis and testing For chemoinformaticians, the task is to develop and utilize various computer programs to evaluate a very large number of chemical compounds and recommend the most promising ones for bench medicinal chemists. This process can be called virtual 

The advantage of using QSAR models for database mining is that it affords not only the compounds selection but also quantitative prediction of their activity. For illustration, we shall discuss our recent success in developing validated predictive models of anticonvulsants [51] and their application to the discovery of novel potent compounds by the means of database mining [10]. 

Experimental Validation I As a result of computational screening of the NCI database, 27 compounds were selected as potential anticonvulsant agents and submitted to our experimental collaborators. Of these 27 compounds, our collaborators selected two 

Experimental Validation 11 Mining of the Maybridge database yielded two compounds that were synthesized and sent to NIH for the MBS anticonvulsant test. Very promising and exciting results have also been obtained. One of the compounds (C9) shows moderate anticonvulsant activity of ED50 between 30 and 100 mg kg (in mice), while the other (C8) is a very potent anticonvulsant agent with ED50 of 18 mg kg in mice (ip). In summary, both compounds were found to be very active in both mice and rats (Table 16.2). 

---

Validated QSAR Models as Virtual Screening Tools 445... 

---