Datasets:

zhihuanglab
/

Paladin_TCGA_CPTAC_omics

Tags:

License:

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/hdf5/hdf5.py", line 49, in _split_generators
                  import h5py
              ModuleNotFoundError: No module named 'h5py'
              
              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 71, 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.

Paladin TCGA & CPTAC Spatial Omics Maps

Ready-to-use patch-level and slide-level spatial omics maps inferred by Paladin from TCGA and CPTAC whole-slide images. The released .Paladin.h5 files can be analyzed directly without rerunning WSI inference.

The collection is populated in stages. Check Files and versions for the cohorts currently available.

Spatial multi-omics example

The panels show the H&E WSI, a reference tumor mask, CNV burden, TP53 CNV, DNA-methylation burden, CDX2 spatial RNA, hsa-miR-21-5p miRNA, and PanCK CODEX protein for one TCGA-COAD slide. The visualization is generated directly from a released .Paladin.h5 prediction using the executable GitHub notebook.

CNV, methylation, and miRNA are bulk-trained heads, so their patch maps are model-derived spatial predictions and slide_values remains their primary output. Spatial RNA and CODEX protein are patch-level spatial heads. The maps show model-space predictions rather than assay-native measurements.

Download

Install or update the Hugging Face CLI:

pip install -U huggingface_hub

Download the complete released collection:

hf download zhihuanglab/Paladin_TCGA_CPTAC_omics \
  --repo-type dataset \
  --local-dir Paladin_TCGA_CPTAC_omics

Download only one cohort, for example CPTAC colon adenocarcinoma:

hf download zhihuanglab/Paladin_TCGA_CPTAC_omics \
  --repo-type dataset \
  --include "CPTAC/cptac_coad/*.Paladin.h5" \
  --local-dir Paladin_TCGA_CPTAC_omics

The full collection is large, so cohort- or file-level downloads are recommended for exploratory use.

File organization

CPTAC/<cohort>/<slide>.Paladin.h5
TCGA/<cohort>/<slide>.Paladin.h5

Directories appear as cohorts are released.

HDF5 schema

/coords                                      [N, 2] int64
/tasks/<task>/patch_values                   [N, D] float16
/tasks/<task>/feature_names                  [D] string
/tasks/<task>/slide_values                   [D] float32

Rows in patch_values align with /coords; columns align with feature_names. The seven prediction tasks are vaf_bulk, rnaseq_bulk, cnv_bulk, methy_bulk, mirna_bulk, rnaseq_st, and codex_protein. See the Paladin README for target definitions and feature-name examples.

Intended use and limitations

These files contain model predictions, not measured spatial assays. Outputs must not be interpreted automatically as raw counts, integer copy-number calls, calibrated log2 ratios, methylation beta values, fluorescence intensities, or clinical test results. Independent molecular and pathology validation is required for biological or clinical interpretation.

TCGA and CPTAC source-data terms remain applicable. This dataset-card license does not relicense upstream whole-slide images or other third-party source data.

Acknowledgments

We thank the TCGA Research Network, the CPTAC consortium, all participating patients and institutions, Bioptimus for H-optimus-0, and the Mahmood Lab for HEST.

Downloads last month: 8

Total file size:

1.34 TB