Identification of structural fingerprints for ABCG2 inhibition by using Monte Carlo optimization, Bayesian...

Area of Science:

• Medicinal Chemistry
• Computational Biology
• Pharmacology

Background:

• Breast cancer resistance protein (BCRP) is a key factor in multidrug resistance (MDR) to chemotherapy.
• BCRP is a significant target for developing novel small molecule drugs to combat MDR in breast cancer.

Purpose of the Study:

• To investigate the relationship between molecular structure and ABCG2 inhibition.
• To identify critical structural fingerprints that modulate ABCG2 inhibitory activity.

Main Methods:

• A multi-QSAR approach was employed using diverse datasets of 294 ABCG2 inhibitors.
• Statistical methods included Monte Carlo Optimization, Bayesian classification, and random forest analysis.
• Structural fingerprints were identified and validated across different modeling techniques.

Main Results:

• The best QSAR models demonstrated good predictive performance (e.g., accuracy up to 0.756, balanced accuracy 0.750).
• Key structural features influencing ABCG2 inhibition were identified through various chemometric methods.
• Cross-validation confirmed the robustness and reliability of the identified structural fingerprints.

Conclusions:

• This study provides valuable insights into structural determinants of ABCG2 inhibition.
• The identified structural fingerprints can guide the rational design of novel BCRP inhibitors for breast cancer therapy.
• This research contributes to the development of strategies to overcome multidrug resistance in cancer treatment.