Related Experiment Videos
Predicting ADME properties and side effects: the BioPrint approach.
Cecile M Krejsa1, Dragos Horvath, Sherri L Rogalski
1Cerep, 15318 NE 95th Street, Redmon, WA 98052, USA. ckrejsa@cerep.com
Current Opinion in Drug Discovery & Development
|September 4, 2003
Summary
The BioPrint database enhances drug discovery by providing a large, standardized dataset for computational modeling. This accelerates the prediction of absorption, distribution, metabolism, excretion (ADME), and adverse drug reactions (ADR).
Area of Science:
- Computational chemistry and cheminformatics
- Pharmacology and drug discovery
- Bioinformatics and data science
Background:
- Accurate prediction of drug absorption, distribution, metabolism, excretion (ADME), and adverse drug reactions (ADR) is challenging due to physiological complexity.
- Existing computational modeling approaches are limited by the lack of large, robust, and standardized training datasets.
- The BioPrint database was created to address this gap by systematically profiling marketed drugs and reference compounds.
Purpose of the Study:
- To present the BioPrint database as a comprehensive resource for computational drug discovery.
- To demonstrate how integrated computational chemistry and in vitro data can improve predictive models.
- To discuss the application of in silico methods for predicting clinical liabilities and accelerating drug development.
Main Methods:
- Systematic profiling of marketed drugs and reference compounds to build the BioPrint database.
- Integration of compound structures, molecular descriptors, in vitro ADME/pharmacology profiles, and clinical data (pharmacokinetics, ADRs).
- Development of computational tools utilizing chemical structure and in vitro results as descriptors for predicting in vivo endpoints.
Main Results:
- The BioPrint database contains extensive datasets including compound structures, molecular descriptors, in vitro ADME/pharmacology, and clinical data.
- Computational models are strengthened by incorporating in vitro results, improving predictions of complex in vivo endpoints.
- The platform facilitates the integration of computational chemistry into library design and lead optimization.
Conclusions:
- The BioPrint pharmacoinformatics platform systematically accelerates drug discovery and improves quantitative structure-activity relationships (QSAR).
- Development of robust in vitro/in vivo associations is enabled by the database and associated computational tools.
- The study highlights the importance of training set size/diversity and the utility of in silico methods for predicting clinical issues.