A Tutorial and Use Case Example of the eXtreme Gradient Boosting (XGBoost) Artificial Intelligence Algorithm for Drug...

Area of Science:

• Computational Biology
• Pharmacology
• Data Science

Background:

• Artificial intelligence and machine learning (AI/ML) are increasingly vital in modern drug development.
• A strong grasp of AI/ML principles is essential for selecting appropriate methods.
• Understanding specific algorithms like XGBoost is key for practical application.

Purpose of the Study:

• To provide a tutorial on the concepts and implementation of the eXtreme gradient boosting (XGBoost) algorithm.
• To demonstrate XGBoost's application in classification and regression using clinical trial-like datasets.
• To bridge the gap between theoretical AI/ML concepts and practical coding for drug development.

Main Methods:

• Focus on the eXtreme gradient boosting (XGBoost) algorithm.
• Utilize simple clinical trial-like datasets for classification and regression tasks.
• Emphasize the connection between XGBoost's underlying concepts and its code implementation.

Main Results:

• Readers will gain knowledge of the XGBoost algorithm's principles.
• Readers will learn how to implement XGBoost functions for drug development questions.
• Practical machine learning experience applicable to broader problems will be acquired.

Conclusions:

• This tutorial enhances understanding and practical application of XGBoost in drug discovery.
• It equips researchers with valuable machine learning skills for clinical trial data analysis.
• The acquired knowledge facilitates the use of AI/ML in addressing complex drug development challenges.