HolobiomicsLab single-cell-chromatin-data-handling
Summary
Export peak-by-cell matrices from ArchR processed scATAC-seq projects into formats compatible with trajectory analysis tools such as STREAM. This skill bridges chromatin accessibility data with downstream trajectory inference by standardizing matrix representation and format.
When to use
After calling peaks and annotating cells in an ArchR project, when you need to perform trajectory analysis using STREAM or other external tools that require a peak-by-cell matrix in a specific tabular format (CSV or TSV) rather than native ArchR objects.
When NOT to use
- Input data are already in STREAM-native format or another trajectory tool format; re-exporting will cause redundant processing.
- Peak calls have not yet been performed on the ArchR project; missing peak annotations will result in an empty or malformed matrix.
- Using trajectory tools other than STREAM that have native ArchR support (monocle3, Slingshot); ArchR provides direct integration functions for these.
Inputs
- ArchR project object (processed with peak calls and cell annotations)
- Peak-by-cell accessibility matrix (internal to ArchR project)
Outputs
- STREAM-compatible peak-by-cell matrix file (CSV or TSV format)
- Peak identifiers and cell barcodes in STREAM-expected tabular layout
How to apply
Load a processed ArchR project object containing peak calls and cell annotations. Call the exportPeakMatrixForSTREAM function on the ArchR project to generate a peak-by-cell matrix formatted for STREAM compatibility. Write the resulting matrix to a file in STREAM-compatible format (typically CSV or TSV). The function handles matrix transposition and formatting internally; the user need only specify the output file path. This enables seamless integration with trajectory analysis workflows while preserving the peak-cell accessibility patterns derived from scATAC-seq analysis.
Related tools
- ArchR (scATAC-seq processing, peak calling, and matrix export via exportPeakMatrixForSTREAM function) — https://github.com/GreenleafLab/ArchR
- STREAM (Trajectory analysis tool that accepts the exported peak matrix for cell state inference)
- R (Programming environment for executing ArchR functions and file I/O operations)
Examples
exportPeakMatrixForSTREAM(ArchRProj = proj, outputFile = "peaks_for_stream.csv")
Evaluation signals
- Output file exists and is readable as CSV or TSV with consistent row (peak) and column (cell) structure.
- Peak identifiers and cell barcodes in the exported matrix match those in the original ArchR project.
- Matrix dimensions (peaks × cells) are non-empty and consistent with the peak calls and cell annotations in the input project.
- STREAM successfully ingests the exported matrix without parsing or format errors.
- Accessibility values in the matrix are numeric, non-negative, and fall within the expected range for the scATAC-seq assay (e.g., 0–1 for normalized counts or log-transformed values).
Limitations
- exportPeakMatrixForSTREAM is specific to STREAM compatibility; users requiring export for other trajectory tools must use tool-specific functions or manually format the matrix.
- The export assumes peaks have been called and cells have been annotated in the ArchR project; incomplete or missing annotation data will propagate to the output file.
- No changelog was provided in the source material, limiting visibility into version-specific behavior or recent bug fixes.
Evidence
- [intro] ArchR enables export of peak matrix compatible with STREAM via exportPeakMatrixForSTREAM function: "ArchR now enables export of a peak matrix that is compatible with STREAM! See updates with exportPeakMatrixForSTREAM"
- [other] Workflow involves loading project, calling export function, and writing matrix to file: "Load a processed ArchR project object containing peak calls and cell annotations. 2. Call exportPeakMatrixForSTREAM on the ArchR project to generate a peak-by-cell matrix formatted for STREAM"
- [readme] ArchR is full-featured R package for scATAC-seq: "ArchR is a full-featured R package for processing and analyzing single-cell ATAC-seq data."
- [readme] ArchR supports direct trajectory analysis with monocle3 and Slingshot: "ArchR now directly supports both monocle3 and Slingshot based trajectory analysis!"