Hugging Face's logo

Datasets:
vcc-uba
/
amber_quiet_heron

Tags:
single-cell
perturb-seq
crispri
virtual-cell-challenge
License:
Dataset card Files
xet
 Community
Dataset Viewer
The dataset viewer is not available for this subset.
Cannot get the split names for the config 'default' of the dataset.
Exception:    SplitsNotFoundError
Message:      The split names could not be parsed from the dataset config.
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 286, in get_dataset_config_info
                  for split_generator in builder._split_generators(
                                         ~~~~~~~~~~~~~~~~~~~~~~~~~^
                      StreamingDownloadManager(base_path=builder.base_path, download_config=download_config)
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 80, in _split_generators
                  raise ValueError(
                  ...<2 lines>...
                  )
              ValueError: The TAR archives of the dataset should be in WebDataset format, but the files in the archive don't share the same prefix or the same types.
              
              The above exception was the direct cause of the following exception:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/split_names.py", line 68, in compute_split_names_from_streaming_response
                  for split in get_dataset_split_names(
                               ~~~~~~~~~~~~~~~~~~~~~~~^
                      path=dataset,
                      ^^^^^^^^^^^^^
                      config_name=config,
                      ^^^^^^^^^^^^^^^^^^^
                      token=hf_token,
                      ^^^^^^^^^^^^^^^
                  )
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 340, in get_dataset_split_names
                  info = get_dataset_config_info(
                      path,
                  ...<6 lines>...
                      **config_kwargs,
                  )
                File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 291, in get_dataset_config_info
                  raise SplitsNotFoundError("The split names could not be parsed from the dataset config.") from err
              datasets.inspect.SplitsNotFoundError: The split names could not be parsed from the dataset config.

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

VCC-UBA perturbation data release (2026-10-12)

Raw integer UMI counts for single-target and control-only cells from 33 public Perturb-seq / CROP-seq datasets (20,833,631 cells), harmonised for the vcc-uba training pipeline (2026 Virtual Cell Challenge).

  • core.tar.zst: cell map parquet (one row per cell: dataset, cell line, condition, target gene, guides, control flag, source locator), verify.tsv, HGNC reference.
  • <ID>.tar.zst: per-dataset count store processed/<ID>.h5ad (CSR int32 counts projected onto the 18,533-gene axis of gene_names.csv, zero-padded, with a measured-gene mask) plus sidecars. G37 is split into .part00/.part01 (concatenate before extracting).
  • data_release.tsv: bytes, sha256, cell counts per file.

Download

From a clone of the GitHub repo:

scripts/fetch_data.sh --datasets all            # or e.g. --datasets G05,G18

It downloads, checks sha256, and extracts into data/ (or $VCC_DATA_ROOT). See docs/DATA.md in the GitHub repo.

Provenance and caveats

These are derived copies of public datasets; each remains under its original authors' terms — cite and follow the license of the original study. Counts are as deposited by the authors, except G18 and N10 (re-quantified from SRA with kallisto|bustools, exonic) and G54 (authors' Cell Ranger output with introns). No cell QC filter was applied beyond the authors' own; N10 contains near-empty droplets. Store verification against the sources is sampled (>=500 cells per dataset).

ID cells source
G01 63,410 Cell paper
G04 19,587 paper
G05 169,348 paper
G06 8,843 Genome Research paper
G08 35,601 paper
G12 26,074 paper
G14 2,547,877 Cell paper
G16 98,315 paper
G18 71,544 paper
G20 30,908 paper
G21 1,375 paper
G23 17,076 paper
G25 275,799 Nature paper
G28 74,312 Cell paper
G29 408,429 paper
G30 1,628,476 paper
G36 33,159 Nature paper
G37 7,943,468 preprint; preserve repository noncommercial license
G38 933,848 preprint
G41 1,150,539 preprint
G45 499,998 preprint
G46 861,432 preprint
G47 4,298 Public GEO submitter record; paper NR at cutoff
G50 6,846 Public GEO submitter record; publication NR at cutoff
G53 2,821,016 Nature Biotechnology paper
G54 37,691 paper
G55 28,466 preprint
G57 32,748 Neuron paper
G62 124,963 paper
G63 708,505 paper
G64 2,340 paper
N10 159,178 Science paper
N23 8,162 paper
Downloads last month
420
Total file size:
100 GB