Pressor mechanism evaluation for phytochemical compounds using in silico compound-protein interaction prediction

Area of Science:

• Pharmacology
• Computational Biology
• Bioinformatics

Background:

• Understanding the mechanisms of blood pressure (BP) elevation is crucial for drug development.
• Differentiating pressor mechanisms of various compounds is challenging.
• Compound-protein interactions offer a potential avenue for mechanism elucidation.

Purpose of the Study:

• To develop and validate a method for evaluating pressor mechanisms using compound-protein interactions.
• To classify target proteins involved in BP elevation.
• To explore the potential for classifying pharmacological or toxicological mechanisms of compounds.

Main Methods:

• Collected data on 26 phytochemical components and 46 target proteins related to BP elevation.
• Utilized a random forest model for in silico prediction of compound-protein interactions.
• Employed hierarchical clustering, network visualization, and principal component analysis (PCA).

Main Results:

• Predicted compound-protein interactions and visualized them using heat maps and networks.
• PCA successfully classified pressor targets into three main classes: neurotransmitter receptors, hormone receptors, and monoamine oxidases.
• Identified potential associations between steroid glycosides and hormone receptors.

Conclusions:

• Compound-protein interaction analysis is a viable approach for classifying pharmacological and toxicological mechanisms.
• The developed method can differentiate compounds based on their pressor mechanisms.
• This approach can aid in deducing mechanisms for uncharacterized compounds.