onlineFDR

onlineFDR implements online multiple-hypothesis testing procedures to control the False Discovery Rate (FDR) during sequential analysis of dynamically accumulating P-values.

Key Features:

• Adaptive FDR Control: Implements online algorithms to maintain FDR control as new hypotheses and P-values arrive.
• Sequential Testing Capability: Supports hypothesis testing in an online manner where future P-values and the total number of hypotheses are unknown.
• Javanmard and Montanari Procedures: Implements the online FDR procedures developed by Javanmard and Montanari for adaptive multiple-comparison control.
• R Package Implementation: Provided as an R package for integration into R-based analysis workflows.

Scientific Applications:

• Genomics: Controls FDR in sequential testing of genomic datasets that expand over time.
• Proteomics: Controls FDR in sequential testing of proteomic datasets with continuously accumulating experiments.
• Other evolving biological repositories: Supports FDR control in other areas of biological research where data repositories are continuously expanding.

Methodology:

Leverages the online multiple-comparison procedures of Javanmard and Montanari to dynamically adjust for multiple comparisons as new P-values are observed.