Outputs
Find the Spatial Data Object, per-stage outputs, and provenance files the multiomics workflow writes.
The workflow writes one directory per stage under --outdir, plus the Spatial Data Object (SDO) and pipeline provenance. The SDO is the primary output; the stage directories hold the intermediate artifacts and statistics that produced it.
Directory layout
{outdir}/
├── cells2stats/ Cell segmentation, spatial coordinates, per-cell statistics, QC report
├── bases2fastq/ Demultiplexed FASTQ files, run metrics
├── fastq2parquet_3prime/ 3′ alignment and assignment (runs with 3′ RNA batches only)
├── spatialdata/ The Spatial Data Object ({run_id}.zarr.zip) and its layers
└── pipeline_info/ Provenance, tool versions, execution trace and report
The complete file-level layout and per-file descriptions are in docs/outputs.md in the workflow repository.
| Directory | Contents |
|---|---|
cells2stats/ | Cell segmentation masks, spatial coordinates, per-cell statistics, and the Cells2Stats QC report. |
bases2fastq/ | Demultiplexed FASTQ files and run metrics. |
fastq2parquet_3prime/ | Selected gene metadata, per-sample alignment and RNA metrics, polony-level read assignments, and run-level aggregated QC statistics. Present only for runs with 3′ RNA batches. |
spatialdata/ | The Spatial Data Object as {run_id}.zarr.zip, its layers as separate archives (points-zarr.zip, cell-shapes-zarr.zip, nuclear-shapes-zarr.zip), and cyto.viz, the CytoCanvas™ visualization sidecar. |
pipeline_info/ | Workflow provenance, resolved parameters, per-tool versions, and the Nextflow execution trace and report. |
Spatial Data Object
spatialdata/{run_id}.zarr.zip is the input to the multiomics Python package and to CytoCanvas. Load it in Python as described in Load an SDO.
Inside the store, metadata/inputs/ keeps a copy of the small inputs each stage received — run/ (run parameters, manifest, panel), cells2stats/, bases2fastq/, fastq2parquet_3prime/, and pipeline/Provenance.json, the workflow's own provenance record — so the SDO is self-describing.
Provenance
pipeline_info/Provenance.json has three sections:
| Section | Contents |
|---|---|
Workflow | Workflow name, version, Nextflow version, run name, session ID, command line (with the -params-file value redacted), profile, and the git commit and revision when launched from a repository. |
Tools | One {image, url, tag} entry per pinned container image. |
Params | The full resolved parameter map as a flat name: value object, including defaults and launch overrides. |
Tools and Params are also embedded in fastq2parquet_3prime/RunStats.json, so that file records the parameters that produced its statistics. pipeline_info/software_versions.yml lists the runtime-reported version of every tool, one entry per process.
Statistics
| File | Contents |
|---|---|
cells2stats/RunStats.json | Run-level cell segmentation and assignment statistics from Cells2Stats. |
bases2fastq/RunStats.json | Demultiplexing statistics from Bases2Fastq™. |
fastq2parquet_3prime/RunStats.json | Run summary plus per-sample alignment, assignment, mapping, and cell metrics for the 3′ stage. |