Discovery of Novel Antimalarial Compounds Enabled by QSAR-Based Virtual Screening

Abstract

Quantitative structure–activity relationship (QSAR) models have been developed for a dataset of 3133 compounds defined as either active or inactive against P. falciparum. Since the dataset was strongly biased towards inactive compounds, different sampling approaches were employed to balance the ratio of actives vs. inactives, and models were rigorously validated using both internal and external validation approaches. The balanced accuracy for assessing the antimalarial activities of 70 external compounds was between 87% and 100% depending on the approach used to balance the dataset. Virtual screening of the ChemBridge database using QSAR models identified 176 putative antimalarial compounds that were submitted for experimental validation, along with 42 putative inactives as negative controls. Twenty five (14.2%) computational hits were found to have antimalarial activities with minimal cytotoxicity to mammalian cells, while all 42 putative inactives were confirmed experimentally. Structural inspection of confirmed active hits revealed novel chemical scaffolds, which could be employed as starting points to discover novel antimalarial agents.

Activity cliffs in drug discovery: Dr Jekyll or Mr Hyde?

“…A remedial measure to diminish the loss of applicability domain can be to develop several diverse machine learning models and implement a consensus classifier [67,68]. It is well known that multiple, ensemble or consensus classifiers are effective mainly because they span the decision space because each base classifier covers a different region of the decision space (chemical space or SAR) and the union of all the base classifiers produces a common region that results in a wider coverage of chemical space or applicability domain [67,69].…”

Short Communication: Cheminformatics Analysis to Identify Predictors of Antiviral Drug Penetration into the Female Genital Tract

“…9,10 These techniques have been particularly helpful for ADME/Tox predictions, providing a useful and reliable alternative to in vivo assessments. QSAR modeling has also been helpful in predicting drug distribution/penetration into specific biological compartments such as the blood-brain barrier (BBB); for instance, a highly predictive model of penetration (R 2 = 0.80 in an external validation set of 10 compounds) as well as specific contributors to penetration, such as van der Waals surface area and active transport, was reported recently.…”

Polypharmacology of dopamine D1-like receptor antagonists

“…Based on the predicted activities on the targets and estimated pharmacokinetic profiles of designed multipotent ligands are selected the most promising candidates for further study [8][9][10][11][12].…”