MultiPower

MultiPower estimates statistical power and optimal sample sizes for multi-omics experiments to support experimental design and evaluation of multi-omic datasets.

Key Features:

• Harmonized Figures of Merit (FoM): Provides standardized quality descriptors applicable across genomics, proteomics, and metabolomics to enable cross-platform performance assessment.
• Optimal Sample Size Estimation: Estimates and assesses optimal sample sizes for multi-omics experiments across diverse experimental settings, data types, and sample sizes.
• Graphical Decision-Making Support: Produces graphical outputs to aid interpretation and decision-making in experimental design.
• Integration with MultiML: Incorporates the MultiML algorithm to estimate sample sizes for machine learning classification problems using multi-omic data.

Scientific Applications:

• Multi-omics experimental design: Planning and power calculation for studies integrating genomics, proteomics, and metabolomics measurements.
• Machine learning sample-size planning: Determining sample sizes required for classification tasks on multi-omic datasets using MultiML.
• Cross-platform performance assessment: Comparative evaluation of measurement quality and study power across different omic technologies using harmonized FoMs.

Methodology:

Experimental designs are formulated using harmonized Figures of Merit (FoM), and sample-size estimation for machine learning classification is performed with the MultiML algorithm, with adaptation to different experimental settings and data types.