Information entropy-based classification of triterpenoids and steroids from Ganoderma

Area of Science:

• Natural Product Chemistry
• Phytochemistry
• Computational Chemistry

Background:

• Ganoderma species are rich sources of bioactive triterpenoids and steroids.
• Understanding the structure-activity relationships of these compounds is crucial for drug discovery.
• Existing classification methods may not fully capture the nuances of Ganoderma-derived phytochemicals.

Purpose of the Study:

• To develop a novel periodic classification system for triterpenoid and steroid compounds from Ganoderma.
• To utilize artificial intelligence and information entropy for classifying these natural products.
• To correlate structural features with antioxidant activity for identifying potent compounds.

Main Methods:

• A dataset of 71 triterpenoid and steroid compounds from Ganoderma was compiled.
• Information entropy and artificial intelligence were employed for hierarchical classification.
• Six key structural features were hierarchically utilized in the classification process.
• The classification was validated against existing literature data on bioactivity and antioxidant potency.

Main Results:

• A periodic classification system for Ganoderma triterpenoids and steroids was successfully established.
• Compounds within the same group of the periodic table showed similar antioxidant properties.
• Compounds within the same period exhibited maximum structural resemblance.
• A specific steroid with a conjugated three-ketone and two double bonds demonstrated the highest antioxidant activity.

Conclusions:

• The developed periodic classification effectively categorizes Ganoderma triterpenoids and steroids based on structural and electronic properties.
• This AI-driven approach provides insights into structure-activity relationships, particularly for antioxidant potential.
• The classification system aids in predicting and identifying highly bioactive compounds from natural sources.