Predicting P-Glycoprotein-Mediated Drug Transport
Author Information
Author(s): Bikadi Zsolt, Hazai Istvan, Malik David, Jemnitz Katalin, Veres Zsuzsa, Hari Peter, Ni Zhanglin, Loo Tip W., Clarke David M., Hazai Eszter, Mao Qingcheng
Primary Institution: Virtua Drug Ltd., Budapest, Hungary
Hypothesis
Can a support vector machine (SVM) method accurately predict P-glycoprotein (P-gp) substrates based on known data?
Conclusion
The study developed a predictive model that achieved approximately 80% accuracy in predicting P-gp substrates.
Supporting Evidence
- The SVM method showed a prediction accuracy of approximately 80% on an independent external validation data set of 32 compounds.
- A homology model of human P-gp was constructed based on the X-ray structure of mouse P-gp.
- Molecular docking successfully predicted the geometry of P-gp-ligand complexes.
- The web server developed allows users to predict whether a compound is a P-gp substrate.
Takeaway
Scientists created a computer program that helps figure out if a medicine can be moved by a special protein in our body, which is important for how well the medicine works.
Methodology
The study used a support vector machine (SVM) method and molecular docking to predict P-gp substrates based on a dataset of known substrates and non-substrates.
Potential Biases
Conflicting reports in literature regarding the classification of compounds as P-gp substrates or non-substrates could introduce bias.
Limitations
The model may not accurately predict all P-gp interactions due to potential false positives and the complexity of drug interactions.
Statistical Information
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website