marr

marr implements a nonparametric approach that detects reproducible signals using a maximal rank statistic for high-dimensional biological data.

Key Features:

• Feature-level reproducibility measurement: Measures the reproducibility of features per sample pair in high-dimensional biological replicate experiments.
• Sample pair-level reproducibility measurement: Measures the reproducibility of sample pairs per feature in high-dimensional biological replicate experiments.
• Histogram visualization: Plots histograms of the reproducibility of features per sample pair and sample pairs per feature.
• Threshold selection support: Enables selection of optimal filtering threshold values for identification of reproducible features and sample pairs based on output visualization checks.
• Data subsetting: Provides the subset of data filtered by reproducible features and/or sample pairs.

Scientific Applications:

• Reproducible signal detection: Identifies reproducible signals in high-dimensional biological replicate experiments using a maximal rank statistic.
• Feature filtering: Filters high-dimensional biological data to retain only reproducible features for downstream analysis.
• Sample quality assessment: Evaluates and filters sample pairs based on reproducibility metrics to identify reliable replicates.

Methodology:

marr uses a maximal rank statistic, a nonparametric approach, to detect reproducible signals in high-dimensional biological data.