During the lifespan of a project, researchers have to deal with data management on a daily basis. Active research data management refers to the tasks and tools required to ensure that data, code, and related information remain organized and safely backed up, for research to be reproducible and secure. The key to good active research data management is data documentation.
Data preparation
Formatting
A file format is a standard way to encode data for storage in a computer file. File formats can be proprietary or free and can be unpublished or open.
When selecting file formats, possibly choose formats that are interoperable among various platforms and applications, open, and commonly used by the research community. If data are stored in one format during collection and analysis, and then transferred to another format for preservation, be careful to list out features that may be lost in data conversion.
Metadata
Metadata is “data that provide information about other data” (source: Merriam-Webster). Data documentation and metadata provide essential information about the context, structure, provenance, and content of the data. The goal is to allow others (including your future self) to use and interpret the data. The minimum documentation of a dataset is to describe it within a README file and, if appropriate, a naming convention.
Data generation and documentation
ELNs and LIMS
Electronic laboratory notebooks (ELNs) and laboratory information management systems (LIMS) are applications that allow researchers in a laboratory to track samples and test results. An electronic laboratory notebook (ELN) replicates a digital version of the traditional paper notebook with the advantage of many built-in features.
Data processing and management
Version control
Version control is a method that allows keeping track of file (data and code) changes over time, so that older versions remain accessible in the future.
Processing environment and workflows
A scientific workflow is a formal definition of the research process. In addition to automating tasks, such formalization increases research reproducibility. Workflows are made of a series of computational or data manipulation steps and are machine-readable. Scientific workflow management software allows one to easily manage complex or repetitive operations.
Data storage and security
Storage, back-up and cloud
For general storage, use File Storage, the central storage and backup service by EPFL VPO. It also offers an “object storage” hosted on-site and based on Open Standard S3 protocol: use the XaaS portal to request for buckets.
For storage capacity and help, please refer to your Faculty-IT.
Data sharing and collaboration
Synchronize and share
File synchronization and research data sharing can be done through various platforms, depending on the needs and location of the data and partners. Since cloud-based solutions are often chosen, remember to check for any personal data protection or sensitive data.