Machine Learning Made Easy (MLme): a comprehensive toolkit for machine learning-driven data analysis

Area of Science:

• Bioinformatics
• Computational Biology
• Data Science

Background:

• Machine learning (ML) is crucial for analyzing complex datasets, but developing ML pipelines is time-consuming and requires specialized expertise.
• Current ML tools often demand advanced programming skills and extensive pipeline configuration, hindering research progress.
• Researchers face challenges in effectively utilizing ML due to the complexity and resource-intensive nature of existing solutions.

Purpose of the Study:

• To develop a user-friendly tool, Machine Learning Made Easy (MLme), that streamlines the application of ML in research for classification problems.
• To eliminate the need for extensive coding knowledge and complex pipeline configuration for researchers.
• To provide a comprehensive solution integrating data exploration, automated ML, custom ML, and visualization functionalities.

Main Methods:

• Developed MLme, a novel tool with four integrated functionalities: Data Exploration, AutoML, CustomML, and Visualization.
• Tested MLme rigorously on six diverse datasets with unique characteristics and challenges.
• Utilized MLme's feature selection capabilities to identify significant cell population markers.

Main Results:

• MLme demonstrated promising and consistent performance across all tested datasets, highlighting its versatility and effectiveness.
• The tool successfully streamlined ML pipeline development, reducing the need for extensive coding.
• Identified significant markers for CD8+ naive (BACH2), CD16+ (CD16), and CD14+ (VCAN) cell populations using MLme's feature selection.

Conclusions:

• MLme serves as a valuable resource for researchers, simplifying ML application for insightful data analysis and improved research outcomes.
• The tool effectively addresses the challenges associated with complex coding scripts in ML.
• MLme is accessible with source code and tutorials available online, promoting wider adoption in research.