Predicting Blood-Brain Barrier Permeation of Drug-Like Compounds
Author Information
Author(s): Bolboacă Sorana D., Jäntschi Lorentz
Primary Institution: Iuliu Haţieganu University of Medicine and Pharmacy Cluj-Napoca
Hypothesis
Can a predictivity statistical approach improve the prediction of blood-brain barrier permeation for drug-like compounds?
Conclusion
The predictivity approach effectively characterizes the model for predicting blood-brain barrier permeation and compares it with previously reported models.
Supporting Evidence
- The model achieved an accuracy of 69% in the training set and 73% in the test set.
- The classification model showed better specificity for inactive compounds (~74%) compared to active compounds (~64%).
- The predictive accuracy on the external set was also 73%.
Takeaway
This study helps scientists figure out which drugs can pass through the blood-brain barrier, which is important for treating brain diseases.
Methodology
The study used a multiple linear regression model based on a set of 437 drug-like compounds to predict blood-brain barrier permeability.
Potential Biases
Potential bias due to the selection of compounds and the variability in experimental data.
Limitations
The model's performance may vary based on the distribution of active and inactive compounds in different sets.
Participant Demographics
The study analyzed 437 diverse drug-like compounds, including 122 with measured BBB permeability.
Statistical Information
P-Value
p<0.05
Confidence Interval
[58.53–78.37]
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website