Coming into this project it seems to me that there are certain logical categories of information that metadata is grouped in.  Initially I had seven categories (* below) but there are three others (4/28/15) I can think of and I decided to consolidate two in Qualities:

  • Metadata - this is metadata about this record not the methodology itself, version, publisher, date, etc...
  • Description* - general information
  • Infrastructure - who is responsible for the analysis and their roles, where the analysis was developed/validated etc. (added 01/12/15)
  • Analyte(s)* - identifiers and info about speciation, phase/fraction, state
  • Materials - chemicals, glassware and supplies that are important to the analysis process [new 03-09-15]
  • Sample(s)* - normally only one, what it is, where it came from, location if appropriate
  • SamplePrep* - important info not overly detailed
  • Technique(s)* - normally only one, instrumental/wet chemical, major characteristics (changed from Instrumental)
  • Quality(ies)
    • Metrics* - common analytical metrics that describe an analysis (moved 02/17/15)
    • Statistics - statistical and chemometric tests run on data from the analysis (added 02/17/15)
    • Validation* - how the method was validated, SRMs, spiked samples, QC data etc (moved 02/17/15)
    • Example data - new category with calibration curves, chromatograms, spectra, peak tables (added 02/17/15)
  • Concepts - standard vocabulary terms for concepts in chemical analysis (placeholder only) (01/27/15)
  • Bibliography* - references to procedures/papers (removed - a citation can be added in the description)

Although this may look like a lot of information, in most (if not all) cases are going to use only a small submit of the fields, and even fewer will be 'required'.  It is important here to try and capture all the information users might need/want.