Medicine plays a monumental role in modern society; however, drug discovery is an expensive endeavour. With the advent high throughput assays, frequent hitters pose a significant problem in early-stage drug discovery. In this thesis, we describe three machine learning models developed to identify potential false positives in popular assay systems such luciferase-based assays, AlphaScreen, and GPCR assays. Such models can be used to identify false positives and frequent hitters early in the drug discovery pipeline, thereby reducing the risk of false leads.
«
Medicine plays a monumental role in modern society; however, drug discovery is an expensive endeavour. With the advent high throughput assays, frequent hitters pose a significant problem in early-stage drug discovery. In this thesis, we describe three machine learning models developed to identify potential false positives in popular assay systems such luciferase-based assays, AlphaScreen, and GPCR assays. Such models can be used to identify false positives and frequent hitters early in the drug...
»