MADAM - An open source meta-analysis toolbox for R and Bioconductor.

Area of Science:

• Biomedical research
• Computational biology
• Statistical genetics

Background:

• Meta-analysis is a crucial tool in biomedical research.
• Existing R packages for meta-analysis often lack parallel computing capabilities.
• The development of new computational tools is essential for advancing research methodologies.

Purpose of the Study:

• Introduce MADAM, a novel R and Bioconductor package for meta-analysis.
• Enhance meta-analysis efficiency through parallel computing.
• Provide tools for ensemble meta-analysis and result visualization.

Main Methods:

• Implemented a parallelized meta-analysis method within the MADAM package.
• Integrated functions for combining results from multiple meta-analysis methods (ensemble approach).
• Developed tools for visualizing meta-analysis outcomes and a data generator for testing.

Main Results:

• MADAM supports five distinct meta-analysis methods and three ensemble methods.
• The package facilitates meta-analysis through direct functions and by wrapping existing implementations.
• Includes three functions for result visualization and a mock data generator for method development.

Conclusions:

• MADAM offers a unified R package for diverse meta-analysis methods and data types.
• Compatibility with existing parallel computing infrastructure is achieved via the 'snow' package.
• MADAM is open-source and freely available on CRAN for broad accessibility.