This document it is assumed to be a “living” document. We highly appreciate any comments and suggestions for improvements. What are your experiences with research data managment tasks? Can you provide solutions for specific tasks? We want your feedback to make the document better for you and your colleages. To add your annotation, select some text and then click the on the pop-up menu. To see the annotations of others, click the in the upper right hand corner of the page

Chapter 4 Integration in QMS {qms}

4.1 Start a new project

Before starting with a new project, perform the following actions that are detailed in the following sections:

At the start of a research project:

At the start of a project (or if an employee or trainee enters the project):

  • Give an introduction to our research data management as described in this document.

Regularly during the project:

  • Check if the folder structure within your project’s rawdata subfolder still complies with the rawdata folder structure and clean the structure, if not.

Project and owner identifiers

Documents and data that are collected or created during the lifetime of a project will be stored in the form of electronic files on the institute’s file server or on the personal computers of the employees. We aim at being able to clearly identify the related project and the owner of the data for any file or folder that resides on the institute’s file system.

Therefore, we recommend to use unique identifiers, either as (part of) the name of the file or as (part of) the name of the folder or parent folder that the file resides in. The usage of a consistent project identifier becomes particularly important when - as defined we suggest for a good data workflow, different types of data and documents of the project are stored at different places on the file system.

Together with the team members, the project leader is requested to define the project identifier at the start of the project. Whenever the belonging of a file or folder to the project is to be indicated, only this “official” identifier in the exact spelling that was defined, has to be used.

The project identifier should

  • be unique within all identifiers of current or former projects that have ever been carried out at the institute,

  • be as short as possible but as long as required to be meaningful and easy to remember.

In order to minimise the risk of spelling mistakes and to avoid errors in automated processing we require the identifier to

  • start with a lowercase letter,
  • consist of only lowercase letters (a-z), digits (0-9) or the hyphen (-),
  • end with a lowercase letter or digit,
  • consist of at least four and at last 16 letters,
  • not contain two hyphens in sequence (--).

Examples for valid project identifiers are:

  • sema-berlin-2
  • aquanes
  • ist4r

Invalid project identifiers are:

  • sema-berlin_2 (has underscore)
  • Aquanes (has uppercase letters)
  • 4you (starts with a digit)
  • abc (does not have at least four letters)
  • this-project-name-is-too-long (has more than 16 letters)

The project identifier is not to be confused with the project acronym or project title as it has been aggreed on by the project consortium during the project development.

We propose to document the choice of project identifier, project acronym and project title in a project’s metadata file. This will be described in another chapter of this document.

For data that are received from external partners or organisations we recommend to indicate the origin of the data by means of another type of identifier in the names of the corresponding files and folders. The same strict naming convention as for the project identifiers should be applied to these institutional identifiers that are to be documented in another special metadata file. At the start of a project it should be checked if the organisations that are expected to receive data from are listed in this metadata file. If not, the metadata file is to be extended as necessary.

Data Management Plan

Data Management Plans (DMP) document the type of data that is expected to be acquired or generating during the life time of a project. The DMP will show you and your colleagues how this data is described, analyzed and stored and how you intend to share and preserve the data after the project has ended. The DMP will help you to formalize data handling processes and discloses potential weaknesses in your project. The overall goal of a DMP is to comply with the FAIR principles that will make your data:

  • Findable
  • Accessible
  • Interoperable
  • Reusable

The DMP gives information on how you intend to make the data findable with metadata and standard identification mechanism (e.g. a DOI). This is achieved through e.g. repositories and usage of metadata standards. The DMP also contains adequate documentation and necessary tools needed to access the data. You are aware that the DMP needs to balance openness on one hand and protection of scientific information, commercialisation and Intellectual Property Rights (IPR) on the other hand. If certain datasets cannot be shared, provide a clear reason why.

Interoperability enables data exchange and re-use between researchers and across institutions and also requires standardised metadata and methodologies. Reusability of data requires the data to be tagged with a license that clarifies how data can be re-used.

Please note that data management is a dynamic process, hence the DMP requires regular adaptations during the course of a project. It is therefore strongly recommended to announce a data manager who takes over responsibility for any data management related issues. Please use DMP online to create the DMP. DMP online is a tool with a lot of tips and links that will construct your DMP according to the funder requirements. It will guide you through the whole DMP process in a threefold process (initial, detailed and the final DMP).

The costs for data management are eligible for most funders. The write up of an initial DMP will cost you between 1 hour and 1 day of work, depending on nature of your project and the data that is acquired.

Data should normally be provided in a non-proprietary format (for example .csv Your organisation is coordinating a proposal for a EU H2020 call and requires a DMP. You can inform the consortium members that each member has opt-out possibilities at application phase, during grant agreement preparation and after signature of the grant agreement.

Look through a (Research) Data Management Planning (DMP) checklist, e.g. here: DMP checklist

If the project’s funding organisation demands a Data Management Plan, use the template that is provided by the funder. You find different DMP-templates (e.g. that required for European Horizon 2020 projects) here: DMP Online.

Responsibilities of the Data Manager

The data manager is responsible for:

  • saving raw data in the rawdata network folder (see below). The data manager gets a special login account with write-access to the rawdata folder. The login information can be got from Michael Rustler or Hauke Sonnenberg, the members of the FAKIN project team.

  • regularly checking if both the file and folder names as well as the folder structure comply with the best practices described in this document and with the naming conventions aggreed on at the start of the project. The data manager curator checks if metadata files are available and up-to-date.

Setup folder structures

At the start of the project, define and create folder structures for the three areas rawdata,processing, results that we propose in this document.

Create the folders and provide a metadata file (see below) describing their meaning. In addition to the general recommendations given in this document, create and document naming conventions for files and folders for your project.