Documentation

Systematic documentation is vital.

Systematic documentation is crucial for data to be used and interpreted later - and for the results to be shared and reproduced. The documentation process should be standardized and possibly automated as much as possible.

Documentation, which should always follow original or raw data, is:

  • log files from measuring instruments and software
  • lab note books - like Electronic Lab Notebooks (ELN)
  • ReadMe files -  describing folder contents, file structures and naming principles
  • scripts
  • sample IDs with description of the samples
  • references to the background literature.

Documentation also includes description of the process, how is data collected, which software is used to read data, metadata, ie. quantities, formats, etc., as well as description of origin/provenance, historical report on usage and changes to data sets.

Metadata
All relevant metadata should be embedded in the respective data set or file and enable other users to fully understand the format and content. Metadata should be self-explanatory.

Descriptive metadata refer to the content of the data set like place and time of an experiment, conditions in the laboratory, titles of columns and rows in a table, parameters, definition of variables, units of measurements, assumptions used, etc. In many research areas, there are well-established and internationally recognized metadata standards. Examples are listed by the Research Data Alliance.

Bibliographic metadata contain author, date, version, links to other data sets or publications, information about the software required to read the data, etc.

Technical metadata are usually generated automatically, like file type, file size, hash values, etc.

Electronic Lab Notebook - ELN

ELN is software that basically has an interface as a page in a paper lab book. In the electronic book you can enter protocols, observations, notes, pictures, drawings, links to other protocols, etc. Some of the commercial applications include a Laboratory Information Management System (LIMS), where you can link your experiments to archived data, to other samples and equipment.

ELN-software, examples:

Metadata

Metadata is additional information that describes a file or an object.

Where applicable, standard metadata schemes should be used, e.g. the DataCite Metadata Schema 4.0 or Dublin Core.