Related Experiment Videos
Correlation of blood-brain penetration using structural descriptors.
Alan R Katritzky1, Minati Kuanar, Svetoslav Slavov
1Center for Heterocyclic Compounds, Department of Chemistry, University of Florida, Gainesville, FL 32611, USA. katritzky@chem.ufl.edu
Bioorganic & Medicinal Chemistry
|May 16, 2006
Summary
This study developed quantitative structure-activity relationship (QSAR) models to predict drug penetration into the brain. These models accurately forecast blood-brain barrier (BBB) penetration for new drug candidates.
Area of Science:
- Medicinal Chemistry
- Computational Chemistry
- Pharmacokinetics
Background:
- Predicting blood-brain barrier (BBB) penetration is crucial for developing effective central nervous system (CNS) drugs.
- Experimental determination of BBB partitioning is resource-intensive.
- Quantitative Structure-Activity Relationship (QSAR) models offer a computational approach to predict drug properties.
Purpose of the Study:
- To develop and validate QSAR models for predicting blood-brain barrier penetration (logBB).
- To correlate experimentally determined logBB values with computed molecular descriptors.
- To compare the performance of models based on molecular versus fragment descriptors.
Main Methods:
- Utilized CODESSA-PRO and ISIDA software for descriptor calculation.
- Developed QSAR models using molecular and fragment descriptors for 113 drug molecules.
- Validated models using an external test set of 40 CNS-active drugs.
Main Results:
- A five-descriptor linear model from CODESSA-PRO achieved R²=0.781 and s=0.123.
- An ISIDA 'consensus model' yielded superior results with R²=0.872 and s=0.047.
- Both models demonstrated successful prediction accuracy on the external validation set.
Conclusions:
- QSAR models based on molecular and fragment descriptors can reliably predict drug BBB penetration.
- The ISIDA 'consensus model' showed higher predictive power.
- These validated models can aid in the rational design of CNS-penetrant drugs.