{
  "dataset": "FORESEE / Walker 2022 colorectal cancer cfDNA input-WGS fragmentomics",
  "bioproject": "PRJNA755688",
  "sra_study": "SRP332946",
  "zenodo_record": "5170265",
  "zenodo_url": "https://zenodo.org/records/5170265",
  "paper_url": "https://www.nature.com/articles/s41598-022-20975-1",
  "paper_doi": "10.1038/s41598-022-20975-1",
  "license": "CC BY 4.0 for the Zenodo processed-data record",
  "input_library": "ordinary paired-end cfDNA WGS input-control library paired with the study's 5hmC assay",
  "files": [
    {
      "name": "FORESEE_sample_description.tsv",
      "bytes": 538010,
      "md5": "f4d1f35bf72f9d8a52a8a5b1b8c3d2c2",
      "sha256": "cd5c1c32ece6748352f2b3bbb043ad97efa4af9cc146a251b2d0958afcba0e9d",
      "url": "https://zenodo.org/api/records/5170265/files/FORESEE_sample_description.tsv/content",
      "status": "downloaded_and_verified"
    },
    {
      "name": "CRC_cegxdelfi_znorm_training_matrix_v1.tsv",
      "bytes": 6557369,
      "md5": "39778142f09306694bd5ca9e526029ce",
      "sha256": "c966445a2ba92db561b6beab563f9bb24e8c55ea7644003bd143ba77669869b0",
      "url": "https://zenodo.org/api/records/5170265/files/CRC_cegxdelfi_znorm_training_matrix_v1.tsv/content",
      "delimiter_detected": "tab",
      "status": "downloaded_and_verified"
    },
    {
      "name": "CRC_cegxdelfi_znorm_validation_matrix_v1.tsv",
      "bytes": 3594129,
      "md5": "637709de6a45c668e7fb4a072dfc3788",
      "sha256": "3eaaf2add7be7e9a7921c52864d54d9df0d8e6d93665deab71c44028f08853ac",
      "url": "https://zenodo.org/api/records/5170265/files/CRC_cegxdelfi_znorm_validation_matrix_v1.tsv/content",
      "delimiter_detected": "comma despite .tsv extension",
      "status": "downloaded_and_verified"
    }
  ],
  "split_warning": "The metadata description field says Training sample for all 1,392 rows. This analysis determines the split from membership in the published training and validation matrices and rejects sample overlap.",
  "raw_pilot_manifest": "real_data/manifests/foresee_raw_pilot.csv",
  "raw_pilot_estimated_sra_lite_gib": 59.4,
  "raw_pilot_estimate_note": "Planning estimate from public SRA run-size metadata; the 16 raw runs have not been downloaded or analyzed in this package.",
  "analysis_boundary": "The empirical benchmark uses the authors' processed 512-bin within-sample-normalized matrices, not raw reads. It is a post-publication same-study published-partition CRC reanalysis, not blinded, preregistered, independent-site, prospective, or MCED validation."
}