Current computational methods for predicting protein interactions of natural products

Area of Science:

• Drug discovery and development
• Computational chemistry
• Pharmacology

Background:

• Natural products (NPs) are crucial drug sources due to their structural diversity.
• Predicting NP interactions with protein targets is vital but challenging.
• Experimental methods for NP profiling are costly and time-consuming.

Purpose of the Study:

• To review recent advances in computational approaches for predicting drug-target interactions (DTIs).
• To focus on ligand-based, target-based, and hybrid computational methods.
• To highlight the potential and limitations of computational DTI prediction.

Main Methods:

• Review of computational methods for DTI annotation.
• Categorization into ligand-based, target-based, and hybrid approaches.
• Analysis of recent advancements in the field.

Main Results:

• Computational methods offer faster and cheaper alternatives to experimental NP profiling.
• These methods can significantly aid in discovering selective drugs with reduced side effects.
• Various computational strategies exist for predicting unknown DTIs.

Conclusions:

• Computational DTI prediction is a powerful tool for advancing drug discovery.
• Overcoming current limitations is key to realizing the full potential of these methods.
• Future research should focus on refining computational approaches for NP profiling.