domains / families / science / scarf-single-cell
scarf-single-cell
authentic
Analyze single-cell RNA-seq at million-cell scale with Scarf: out-of-core Zarr stores on disk or object storage, provenance-tracked artifacts, audited QC, clustering, markers and donor comparisons.
When to use
- Use when a task involves a Scarf `.zarr` store, `scarf.DataStore`, `ds.pipeline` or a
- Use when converting H5AD, 10x, MTX or Seurat data into a Scarf store, or opening and mounting a
- Use for single-cell RNA-seq analysis with core Scarf: QC with removal audits, HVG, PCA and
- Use when counts are too large for memory or live on object storage, so they must be streamed
- Do not use it for `scarf.agent` (the automated agent package), which this skill does not cover.
When not to use
- Covers core Scarf only. It does not use or describe `scarf.agent`.
- Needs Python 3.12+ and scarf 1.0.0rc17 or newer, which is a pre-release. A bare `scarf[extra]`
- Scarf does not check the study design. Donor and replicate structure, confounding and held-out
- Cluster labels from this workflow are hypotheses backed by marker evidence, not a validated
- Numbers in the reference modules come from the 10x 5K PBMC documentation dataset. They are
- Runtime and memory depend on hardware, storage and data. Measured end-to-end runs are on the
- A DataStore is not thread-safe: use one per process.
- Cytebase datasets need the `cytebase` extra and network access, and a mounted store reads counts
Skill page
Source