MatrixQCvis

MatrixQCvis provides interactive visualization of data quality metrics for quantitative omics data in matrix-like format, enabling detection of low-quality samples, drifts, outliers, and batch effects.

Key Features:

• Per-sample and per-feature quality metrics: Visualizes data quality at both sample and feature levels for comprehensive assessment.
• Multiple visualization types: Generates bar plots, violin plots, mean vs standard deviation plots, MA plots, empirical cumulative distribution function (ECDF) plots, sample distance visualizations, and multiple dimension reduction plots.
• Differential expression analysis: Performs analysis using limma (moderated t-tests) and proDA (Wald tests) packages.
• Bioconductor integration: Builds upon the SummarizedExperiment S4 class for integration into existing workflows.

Scientific Applications:

• Metabolomics quality control: Tailored for assessing data quality in metabolomics mass spectrometry experiments.
• Proteomics quality control: Tailored for assessing data quality in proteomics mass spectrometry experiments.
• General omics data quality assessment: Applicable to any quantitative omics data types representable in a SummarizedExperiment object, including count or intensity matrix data.
• Batch effect detection: Identifies systematic technical variations across experimental batches.
• Outlier identification: Detects anomalous samples or features that deviate from expected patterns.

Methodology:

MatrixQCvis uses limma for moderated t-tests and proDA for Wald tests in differential expression analysis. Quality assessment employs standard statistical visualizations including ECDF plots, MA plots, and dimension reduction methods applied to sample distance matrices.