|
Artificial Intelligence (AI)-Based Predictive 3D QSAR Model to Discover Novel 7-Aza-Indole Analogs as Anti-Dengue Agents
|
|
| Author:
|
, ALI QUSAY KHALID, VASUDEVA RAO AVUPATI , HUSNIZA HUSSAIN
|
| Abstract:
|
Development of atom-based three-dimensional quantitative structure-activity relationship (3D QSAR) modelling studies on anti-dengue compounds using artificial intelligence are very limited. Hence, in this study, a series of 7-aza-indole derivatives with their reported anti-dengue activities were considered as ligand dataset to develop and validate Schrodinger Phase™ atom-based 3D QSAR model. Further, this model was exposed to investigate the relationship between structural features and anti-dengue activities. The established machine learning 3D QSAR model is statistically significant (Model: R2 Training Set = 0.71 Q2 (R2 Test Set) = 0.55) with good predictive power. In addition, combined effects contour maps (blue: positive potential & red: negative potential) of this model were critically analyzed and elucidated the pharmacophore features responsible for the observed anti-dengue activities.
|
| Keyword:
|
Artificial Intelligence, Machine Learning, Schrodinger Phase™, Atom-based 3D QSAR, Pharmacophore, Combined Effects Contour Maps, Anti-Dengue Activity, Drug Discovery
|
| EOI:
|
-
|
| DOI:
|
https://doi.org/10.31838/ijpr/2020.SP1.370
|
| Download:
|
Request For Article
|
|
|