Data Sources

External Data Sources

CPTAC-HNSCC

  • Name: The Clinical Proteomic Tumor Analysis Consortium Head and Neck Squamous Cell Carcinoma Collection
  • Version/Date: Version 14 or 15?
  • URL: https://www.cancerimagingarchive.net/collection/cptac-hnscc/
  • Access Method: NBIA Data Retriever
  • Access Date: 2024-03-29
  • Data Format: DICOM-CT, DICOM-RTSTRUCT
  • Citation: National Cancer Institute Clinical Proteomic Tumor Analysis Consortium (CPTAC). (2018). The Clinical Proteomic Tumor Analysis Consortium Head and Neck Squamous Cell Carcinoma Collection (CPTAC-HNSCC) (Version 19) [dataset]. The Cancer Imaging Archive. https://doi.org/10.7937/k9/tcia.2018.uw45nh81
  • License: TCIA Restricted License
  • Data Types:
    • Images: CT, RTSTRUCT
  • Sample Size Selected: TBD
  • PatientIDs:
    • C3N-00297
    • C3N-00498
    • C3N-00828
    • C3N-01620
    • C3N-01752
    • C3N-01754
    • C3N-01757
    • C3N-01943
    • C3N-01948

Crowds-Cure-2018

  • Name: Crowds Cure Cancer: Data collected at the RSNA 2018 annual meeting
  • Version/Date: Version 1: Updated 2019/05/30
  • URL: https://www.cancerimagingarchive.net/analysis-result/crowds-cure-2018/
  • Access Method: NBIA Data Retriever
  • Access Date: 2025-03-26
  • Data Format: DICOM-SR
  • Citation: Urban, T., Ziegler, E., Pieper, S., Kirby, J., Rukas, D., Beardmore, B., Somarouthu, B., Ozkan, E., Lelis, G., Fevrier-Sullivan, B., Nandekar, S., Beers, A., Jaffe, C., Freymann, J., Clunie, D., Harris, G. J., & Kalpathy-Cramer, J. (2019). Crowds Cure Cancer: Crowdsourced data collected at the RSNA 2018 annual meeting [Data set]. The Cancer Imaging Archive. https://doi.org/10.7937/TCIA.2019.yk0gm1eb
  • License: CC BY 3.0
  • Data Types:
    • Annotations: Structured Reports containing 2D RECIST measurements
  • Sample Size Selected: TBD

Overview

This section should document all data sources used in your project. Proper documentation ensures reproducibility and helps others understand your research methodology.

How to Document Your Data

For each data source, include the following information:

1. External Data Sources

  • Name: Official name of the dataset
  • Version/Date: Version number or access date
  • URL: Link to the data source
  • Access Method: How the data was obtained (direct download, API, etc.)
  • Access Date: When the data was accessed/retrieved
  • Data Format: Format of the data (FASTQ, DICOM, CSV, etc.)
  • Citation: Proper academic citation if applicable
  • License: Usage restrictions and attribution requirements

Example:

## TCGA RNA-Seq Data

- **Name**: The Cancer Genome Atlas RNA-Seq Data
- **Version**: Data release 28.0 - March 2021
- **URL**: https://portal.gdc.cancer.gov/
- **Access Method**: GDC Data Transfer Tool
- **Access Date**: 2021-03-15
- **Citation**: The Cancer Genome Atlas Network. (2012). Comprehensive molecular portraits of human breast tumours. Nature, 490(7418), 61-70.
- **License**: [NIH Genomic Data Sharing Policy](https://sharing.nih.gov/genomic-data-sharing-policy)

2. Internal/Generated Data

  • Name: Descriptive name of the dataset
  • Creation Date: When the data was generated
  • Creation Method: Brief description of how the data was created
  • Input Data: What source data was used
  • Processing Scripts: References to scripts/Github Repo used to generate this data

Example:

## Processed RNA-Seq Data
- **Name**: Processed RNA-Seq Data for TCGA-BRCA
- **Creation Date**: 2021-04-01
- **Creation Method**: Processed using kallisto and DESeq2
- **Input Data**: FASTQ Data obtained from the SRA database
- **Processing Scripts**: [GitHub Repo](https://github.com/tcga-brca-rnaseq)

3. Data Dictionary

For complex datasets, include a data dictionary that explains:

Column Name Data Type Description Units Possible Values
patient_id string Unique patient identifier N/A TCGA-XX-XXXX format
age integer Patient age at diagnosis years 18-100
expression float Gene expression value TPM Any positive value

Best Practices

  • Store raw data in data/rawdata/ and never modify it
  • Store processed data in data/procdata/ and all code used to generate it should be in workflow/scripts/
  • Document all processing steps
  • Track data provenance (where data came from and how it was modified)
  • Respect data usage agreements and licenses! This is especially important for data that should not be shared publicly