Data analysis and interpretation is an important and challenging part of the discovery process, due to the large volumes of in vitro/in vivo data generated from pharmacology and high throughput pharmaceutical-profiling programs. The goals of data analysis and interpretation for pharmaceutical profiling are:
1. Model building: correlation between different properties of a drug-candidate series; making useful predictions.
2. Problem solving: diagnosing and forecasting the issues/liabilities and advantages for each series; correlating these to the underlying properties that are responsible.
3. Decision making: providing data and interpretations for teams to make go/no go decisions and to plan optimization work to improve series liabilities.
4. Result presentation: communicating and relating the data in a format that medicinal chemists and pharmacologists can readily grasp and incorporate.
Caldwell  describes a useful method of data analysis and presentation of multiple variable data sets to provide discovery teams with predictions of oral bioavailability. By ranking the compounds as high, intermediate, and low for in vitro solubility, acidic stability (to simulate stomach), hepatocyte stability (to simulate liver), and Caco-2 permeability, reliable early predictions can be made for in vivo oral bioavailability.
Was this article helpful?