BER requires data management and data sharing from DOE-funded work. A system-centralized platform is used by CBI personnel to comply with both DOE and CBI internal requirements. It is expected that all Principal Investigators (PI) store key datasets in an appropriate format (see below) and that data is made available upon publication as data DOIs. LabKey is available to all CBI researchers as a data storage platform. Publication and sharing approaches such as publication and community repositories are allowable if documented. Documentation is satisfied by including a hyperlink to an external resource within a wiki page within a PI’s PUBLIC_DATA folder.
Data organization and Data levels
Data from each CBI associated PI should be stored as indicated below. LabKey contains folders for each PI that follow this format for storage. CBI aspires to follow the FAIR (Findable, Accessible, Interoperable, Reusable) principles for data management. To achieve this objective, CBI assigns different levels to different types of data. For this we take inspiration from NASA’s Earth Data Open access system. (EOSDIS)
In general, raw data, e.g. data that comes directly from a scientific instrument, is considered Level 0. Level 0 data is not expected to be FAIR.
Data you acquire from a third party, e.g. a lab, a notebook, manually entered into a spreadsheet is Level 1 – this data does not have to be FAIR.
Level 0 and Level 1 data are considered CBI business sensitive. This data should not be propagated and re-shared outside of LabKey permission lists.
Data at Levels 0 and 1 are controlled by the PI’s of each laboratory within CBI. Access to these folders can be granted only by the PI or their designated administrator.
While we actively encourage PI’s to submit all Level 0 and Level 1 data that is generated within CBI to the LabKey server for archive and retrieval purposes, we recognize that not all data is “worthy” of uploading. For instance, one-off experiments for calibration purposes, early experimentation, etc. that will never be used in publication and will not be shared are not required to be uploaded. Large experiments that are done using resources outside of a PI’s individual laboratory should always be uploaded and archived. In general, this should be done before the end of the fiscal year in which the experiments were conducted. Data at Level 2 is a candidate for publication.
Level 2 data is your calculated data product. This should be reported in FAIR terms. Level 2 data is expected to conform to normative data standards for whatever data type is being considered.
Data at Level 2 and above should be considered for publication as a data DOI.
Reference CBI’s publication authorship criteria guidelines for inclusion of data generators as co-authors, acknowledgements of citations.
For each datum, your FAIR data product should answer these questions:
- What is the measurement? (This should be a CBI FAIR term)
- What is the unit? (This should come from the FAIR units table)
- What is the uncertainty? (Uncertainty on a digital instrument is typically the lowest value that the instrument can read. Eg. a scale which measure mass as 10.23 kilograms, the uncertainty is +_ 0.3. For calculated values, the uncertainty is the last reported significant digit.
- What is the value? (This is the value of the actual measurement.)
The CBI data management team can help with the transition of data from Levels 0 and Level 1 to Levels 2 and above. We can also assist in obtaining a DOI for publication quality data sets. Contact: Stan Martin for assistance with DOI’s and Level 2 data management.