dimensionality-reduction

To visualize high-dimensional RNA-seq data, this tool reduces numeric expression matrices to 2D or 3D scatter plots using PCA, UMAP, or t-SNE. It outputs projection files and publication-ready plots to reveal sample clustering and patterns.

Yes, you can use t-SNE and UMAP for methylation data visualization. The skill accepts numeric matrices from methylation datasets and applies dimensionality reduction to generate 2D or 3D representations for intuitive exploration.

The best way to check batch effects in multi-omics experiments is applying PCA or UMAP dimensionality reduction. By coloring projections with sample metadata, you can visually assess clustering patterns and identify potential batch outliers.

Sample metadata is optional for generating publication-ready proteomics plots. Providing metadata enables metadata-based coloring to highlight treatment versus control patterns, while the core dimensionality reduction works on a numeric matrix alone.

Yes, this dimensionality reduction tool works with numpy and pandas matrices. It accepts a numeric matrix as input, leveraging scipy and matplotlib dependencies to compute projections and render 2D or 3D visualizations.

Use PCA versus UMAP for clustering analysis based on your goals: PCA provides explained variance and loadings for linear structure, while UMAP preserves local neighborhoods for non-linear clustering. Both generate 2D or 3D projection files.