Optimization to the Phellinus experimental environment based on classification forecasting method.

Area of Science:

• Mycology
• Biotechnology
• Pharmacology

Background:

• Phellinus fungus is a key component in anti-cancer drug development.
• Optimizing Phellinus cultivation is crucial for reliable drug production.
• Previous optimization methods were limited by data scope and experimental experience.

Purpose of the Study:

• To establish optimized culture conditions for Phellinus production using a comprehensive dataset.
• To develop predictive models for Phellinus yield based on culture parameters.
• To enhance the efficiency and yield of Phellinus cultivation for pharmaceutical applications.

Main Methods:

• Collected extensive experimental data on Phellinus culture parameters (inoculum size, pH, temperature, etc.).
• Developed a classification model to distinguish high-yield from low-yield Phellinus cultures.
• Utilized a BP neural network for predicting yield from high-yield data.
• Employed a Genetic Algorithm (GA) to identify optimal culture conditions.

Main Results:

• Achieved a classification accuracy exceeding 90% for high-yield Phellinus cultures.
• The optimized conditions resulted in a slight increase in Phellinus yield compared to previous methods.
• Successfully identified optimal combinations of culture parameters for enhanced Phellinus production.

Conclusions:

• Machine learning models, including BP neural networks and GA, are effective for optimizing fungal cultivation.
• The developed models provide a robust framework for predicting and enhancing Phellinus yield.
• This approach facilitates more efficient and scalable production of Phellinus for anti-cancer drug research.