Quantitative structure-activity relationship analysis of β-amyloid aggregation inhibitors

Area of Science:

• Medicinal Chemistry
• Computational Chemistry
• Neuroscience

Background:

• Alzheimer's disease is characterized by the aggregation of beta-amyloid peptides.
• Inhibiting this aggregation is a key therapeutic strategy.
• Developing effective inhibitors requires efficient screening methods.

Purpose of the Study:

• To develop and validate quantitative structure-activity relationship (QSAR) models for predicting beta-amyloid aggregation inhibitors.
• To identify key chemical features that contribute to inhibition activity.
• To establish computational tools for accelerating the discovery of novel Alzheimer's drug candidates.

Main Methods:

• Collected a dataset of 80 small molecules with known inhibition levels.
• Developed two QSAR models: a Bayesian model and a decision tree model.
• Validated model predictions using in vitro experiments.

Main Results:

• The Bayesian model achieved 87% accuracy on training and test sets.
• The decision tree model achieved 89% and 93% accuracy on training and test sets, respectively.
• Identified key chemical features like electro-topological state of carbonyl groups, AlogP, and hydrogen bond donors as crucial for inhibition.

Conclusions:

• Developed accurate computational models for predicting beta-amyloid aggregation inhibitors.
• Demonstrated the feasibility of these models for rapid drug candidate screening.
• The models can aid in the rational design of novel therapeutics for Alzheimer's disease.