multiome-data-ingestion-paired-modalities

Multiome Data Ingestion for Paired Modalities

Summary

Ingest and align paired scATAC-seq chromatin accessibility and scRNA-seq gene expression data into a unified ArchR project object, establishing the foundation for joint downstream analysis. This skill bridges two single-cell modalities by registering feature matrices and gene expression data in the same coordinate space before dimensionality reduction.

When to use

You have independently generated or received both scATAC-seq peak count matrices and scRNA-seq gene expression matrices from the same set of cells (multiome experiment), and you need to perform joint analysis such as co-clustering, trajectory inference, or regulatory inference that requires both accessibility and expression signals in the same reduced-dimension space.

When NOT to use

• Input cells come from different experiments or individuals — alignment across modalities requires shared cell identity
• Gene expression matrix is already a pre-computed embedding or dimensionality-reduced object rather than a raw/normalized count matrix
• You only have data from one modality (scATAC-seq or scRNA-seq alone) — this skill specifically requires paired data

Inputs

• scATAC-seq peak count matrix (cells × peaks)
• scATAC-seq cell metadata (barcodes, cluster assignments, quality metrics)
• scRNA-seq gene expression matrix (cells × genes, typically log-normalized counts)
• scRNA-seq cell barcodes or metadata aligned to scATAC-seq cells

Outputs

• ArchR project object with both scATAC-seq and scRNA-seq data registered
• Joint reduced-dimension embedding (addCombinedDims output)
• Integrated latent semantic indexing model across both modalities

How to apply

Begin by loading the scATAC-seq peak matrix and associated metadata, then call importFeatureMatrix to register the peak feature matrix into an ArchR project object. Next, load the scRNA-seq gene expression matrix (as a feature matrix) and call addGeneExpressionMatrix to append gene expression data to the same project, aligning cells across modalities by cell barcode. Execute addIterativeLSI on the combined project to jointly compute latent semantic indexing over both accessibility peaks and gene expression counts. Finally, call addCombinedDims to generate a unified reduced-dimension embedding that integrates both scATAC-seq and scRNA-seq signal into a single coordinate space suitable for downstream analysis.