This dataset contains the processed reference dataset from the HeOrganAtlas dataset for Marrow tissue. It has been preprocessed to include log-normalized counts, specific metadata columns, and PCA, t-SNE, and UMAP results.

reference_data

Format

An object of class SingleCellExperiment with 392 rows and 1500 columns.

Source

The HeOrganAtlas dataset, available through the scRNAseq package.

Details

This dataset underwent the following steps:

  • Loads the HeOrganAtlas dataset specifically for Marrow tissue from the scRNAseq package.

  • Divides the loaded dataset into a reference dataset used for downstream analysis.

  • Performs log normalization on the reference dataset using the function logNormCounts from the scuttle package.

  • Selects the column expert_annotation) from the cell metadata for downstream analysis.

  • Selects highly variable genes (HVGs) using the function getTopHVGs from the scran package on the reference dataset.

  • Performs Principal Component Analysis (PCA) on the reference dataset using the function runPCA from the scater package.

  • Performs t-Distributed Stochastic Neighbor Embedding (t-SNE) on the reference dataset using the function runTSNE from the scater package.

  • Performs Uniform Manifold Approximation and Projection (UMAP) on the reference dataset using the function runUMAP from the scater package.

References

He, et al. (2020). HeOrganAtlas: a comprehensive human organ atlas based on single-cell RNA sequencing.

See also

Use data("reference_data") to load and access the resulting reference dataset.

Examples

# Load and explore the reference dataset
data("reference_data")