Related Experiment Videos
Modeling blood-brain barrier partitioning using the electrotopological state.
Kimberly Rose1, Lowell H Hall, Lemont B Kier
1Department of Chemistry, Eastern nazarene College, Quincy, Massacusetts 02170, USA.
Journal of Chemical Information and Computer Sciences
|June 28, 2002
Summary
A new QSAR model predicts blood-brain barrier (BBB) partitioning using molecular structure. It identifies key descriptors for BBB penetration, enabling rapid logBB estimation for drug discovery.
Area of Science:
- Medicinal Chemistry
- Computational Chemistry
- Pharmacokinetics
Background:
- Modeling blood-brain barrier (BBB) partitioning is crucial for drug development.
- Accurate prediction of drug entry into the brain is a significant challenge.
Purpose of the Study:
- To develop a Quantitative Structure-Activity Relationship (QSAR) model for predicting in vivo blood-brain barrier partitioning.
- To identify molecular descriptors that influence BBB penetration.
Main Methods:
- Utilized topological molecular structure representations.
- Developed a QSAR model based on three specific structure descriptors: HS(T)(HBd), HS(T)(arom), and d(2)chi(v).
- Validated the model using external test sets, cross-validation, and prediction on a large drug dataset.
Main Results:
- The QSAR model achieved good predictive performance with low Mean Absolute Error (MAE) and Root Mean Square (rms) values.
- External validation and cross-validation confirmed the model's robustness.
- The model accurately predicted logBB values for a large dataset of drugs and drug-like compounds.
Conclusions:
- The developed QSAR model provides a fast and reliable method for estimating blood-brain barrier partitioning (logBB).
- Key molecular features influencing BBB penetration were identified: presence of aromatic groups, limited hydrogen bond donors, and less branched structures.
- The model's efficiency, due to the exclusion of 3D information, makes it valuable for drug discovery and lead optimization.