NIH | National Cancer Institute | NCI Wiki  

Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: For DSRMWS-1975: Clarified that training is only for curators.

...

CBIIT’s mission is to provide and advocate for the appropriate use of data science, informatics, and information technology (IT) to support and accelerate the NCI Mission to prevent cancer, treat cancer, and improve cancer outcomes. An important role of the NCI Semantic Infrastructure (SI) is to support the NCI research mission through community definition and collection of metadata. Data that have well defined linked metadata can improve the use, interpretation, and reuse of data and the extraction of information and knowledge from these data. Supporting both human readable and machine-readable definitions and metadata has been an important driver for the NCI Semantic Infrastructure. These general metadata characteristics are also among the key principles for data citation , and citation and are noted to enable data access, verifiability, and discoverability.

The primary goals for updating the metadata services are to:

...

The requirements outlined below are described in more detail in the DRAFT Extended MDR RequirementsV10.912.docx (the Extended MDR Requirements is a more descriptive document, adn and therefore lengthy.   We are still working to make a more stakeholder readable version of it).  This document was frozen in June 2016, and provided input for the EPLC documents and the SOW.

MDR Roadmap - Next Steps:

Through an RFI process and research, we identified over 13 potential solutions and partial solutions. We created a requirements spreadsheet and engaged with each of the potential providers to request a self-assessment of their capabilities versus NCI requirements.  We mapped all the results into one document and created a quantitative score for each one based on the number of requirements they could meet. We then organized the requirements by highest priority and re-scored the responses, narrowing the possible choices to the top 5.  We requested product demonstration and information sessions with each of the top 5.  The two possible solutions who held position 4 and 5 respectively turned out not to be commercially available and thus dropped from the list.  We have selected the top 3, scoring 97%, 75%, and 58%, for a 45-day software evaluation for usability testing, and also decided to test the NLM system for comparative purposes. We have developed an "Evaluation Scorecard" to be used in recording results from testing the features that are the most commonly used. The requirements matrix and evaluation scorecard are available upon request from the government sponsor.

We are currently developing two possible solutions for current and future customers to supplement and leverage metadata and models. 

  • Data Mapping and Transformation Tool: Ptolemy.V - Ordinal Data
    • A translation engine that leverages NCI CDEs will create a new file by copying source data you have registered in the tool, and converting it to the common standard format using CDEs. Through a mapping step you can create one or more translations for your data, or by reusing translations that have already been created by other users, you can create new composite, integrated data tables using data from a variety of source data. You can access your tables directly in the repository or you can download/export the data as a CSV file.
    • The project is being managed by Leidos.
  • Metadata Template Builder: CEDAR  - Stanford
    • A web based tool that uses NCI CDEs and leverages open terminologies in BioPortal to create a metadata template to collect and validate data.
    • The project is being managed by Leidos. 


SI Data Semantics Outline

...

- provide a way to record and maintain structured content with minimal effort and training for curators  

3.1.1.4   Download

- provide a human accessible mechanism for getting content out of the repository to meet a variety of end user stories

...

3.1.2.9   Semantic Web Metadata Services (Gilberto FaragosoFragoso)

- The goal of these services are to provide the abilty ability to leverage the semantics of CDEs to explore the meaning of the CDE, the meaning of directly related data, discovery of related data elements, data, and information using semantic web technologies.

3.1.2.10 Community Portals

- Provide the ability to organize content by communities, where CDEs, measures, CRFs and other related information about how to use the standards can be found

3.1.3    New API Services

...

Appendix A – Reporting and content quality

The above outline is presented in full detail in the following document (draft): Extended MDR RequirementsV10.9.docx