Research Data Management lecture for the Summer school of multi-omics (UFZ) - NFDI4Microbiota

Research Data Management (RDM) Martin Bole Prepared in accordance with NFDI4Microbiota, funded by DFG Source material: NFDI4Microbiota - Knowledge Base (CC-BY) Contents 1. Definition of Research Data and Research Data Management (RMD) 2. FAIR data principles 3. Data Management Plans (DMPs) 2 What is Research Data Management? Research Data Management (RDM) “A series of measures that need to be taken during a research project in order to obtain high-quality data (whether produced or reused) ¹ , make data ⁽ ⁾ available and usable over the long-term ² , and make research findings ⁽ ⁾ reproducible beyond the research project ³ “⁽ ⁾ 3 Research Data Life Cycle ¹ Research Data. https://rfii.de/en/topics/#forschungsdaten ² Bres, E., Rudolf, D., Lindstädt, B., & Shutsko, A. (2022). Research Data Management in Medical and Biomedical Sciences. ³ Voight, P., Frericks, S., Lindstädt, B., Shutsko, A., & Vandendorpe, J. (2022). Workshop on Research Data. Research Data Management (RDM) Establishing clear protocols for: -Data collection and generation (SOPs - Standard Operating Procedures) -Data storage (preferably public repositories like ENA, NCBI, DDBJ) -Data analysis (workflow, pipeline, software versioning, models used) -Standardized data formats (FASTA, FASTQ, GenBank, .csv, .tsv, etc.) 4 Research (meta)data Research (meta)data Definition varies across disciplines and research funding agencies “Any information collected, stored, and processed to produce and validate original research results “ (DeWitt Wallace Library; https://libguides.macalester.edu/data1) 5 Data Types Common types of data in microbiology In microbiology, several different types of data exist, all of them important to understand some part of the, dynamic, function, structure of a microbial community or its individual representatives. All of them represent a different layer of information. Common types of data in microbiology: -Genomic sequences, -Amplicon sequences, -Metagenomic sequences, -Metagenome Assembled Genome (MAG) sequences, -(Meta)transcriptomic sequences, -(Meta)proteomic data, -Metabolomic data. Types of metadata Technical metadata refers to the technical part of data generation and processing: ●Instrumentation and Platforms Used ●Sample Collection and Handling Procedures ●Sample Preparation and Extraction Protocols ●Software Tools and Versions Utilized for Data Processing Environmental/Biological metadata refers to the context of sample collection: ●Ecosystem or Habitat Information ●Environmental Conditions (e.g., pH, salinity, temperature) ●Biological Sample Characteristics (e.g., tissue type, diseased status,...) ●Host Organism Metadata Both types of metadata ensures the reproducibility and allows for comparison across different studies. Exercise: Split into groups, think of a project and collect the technical metadata Difference between Technical and Biological metadata Technical metadata Technical metadata in microbiology Is a form of descriptive metadata that refers to the technical part of data generation and processing. This type of metadata ensures the reproducibility of experiments and results and allows for comparison across different studies. Technical metadata includes (but is not limited to): 1. Sequencing Platform Information (Type of sequencing technology, model of sequencer, read length, chemistry, sequencing kit) 2. Sample Collection and Processing (When and how the samples were collected, stored, treatments applied to samples before processing) 3. Nucleic Acid Extraction Protocols (Methods that were used for sample extraction, type of extraction kit or chemicals, nucleic acid concentration, purity of extracted material) 4. Quality Control Measures (Methods for assessing data quality and integrity of DNA/RNA, library quality, sequencing read quality) 5. Data Processing and Analysis (Software and algorithm information, version number, parameters settings for sequence alignment, assembly, annotation, statistical analysis) 6. Primers Used (if applicable) (Information on probes or oligomers, used in sequencing) Data Processing and Analysis Data Processing and Analysis - versions Minimal technical metadata https://anaconda.org/ https://rstudio.github.io/renv/articles/ renv.html Interoperable ¹⁽ ⁾ ¹ Wilkinson, M. D., Dumontier, M., Aalbersberg, I. J. J., Appleton, G., Axton, M., Baak, A., Blomberg, N., Boiten, J.-W., da Silva Santos, L. B., Bourne, P. E., & et al. (2016). The Fair Guiding Principles for Scientific Data Management and Stewardship. Scientific Data, 3(1). https://doi.org/10.1038/sdata.2016.18 ² Action, G. O. D. A. N. (2019). GODAN Action Online Course on Open Data Management in Agriculture and Nutrition (Version v1.0). Zenodo. https://doi.org/10.5281/zenodo.3588148 ³ Luiz Olavo Bonino da Silva Santos, Kees Burger, Rajaram Kaliyaperumal, Mark D. Wilkinson; FAIR Data Point: A FAIR-oriented approach for metadata publication. Data Intelligence 2022; doi: 10.1162/dint_a_00160 What does it mean “To be Interoperable”: 1. (Meta)data uses an accessible, broadly applicable, formal and shared language for knowledge representation (e.g., ontologies, controlled vocabularies, etc.), 2. (Meta)data uses vocabularies that follow the FAIR data principles (e.g., using FAIR Data Point ³ ),⁽ ⁾ 3. (Meta)data indicates and includes a qualified reference to (primary) other (meta)datasets (when used or newly generated dataset builds upon a pre-existing dataset, properly citing all datasets). Dataset is interoperable, when it can be used along other datasets, across different systems, without special effort from the user ² .⁽ ⁾ Reusable (Wilkinson et al., 2016)¹ ¹ Wilkinson, M. D., Dumontier, M., Aalbersberg, I. J. J., Appleton, G., Axton, M., Baak, A., Blomberg, N., Boiten, J.-W., da Silva Santos, L. B., Bourne, P. E., & et al. (2016). The Fair Guiding Principles for Scientific Data Management and Stewardship. Scientific Data, 3(1). https://doi.org/10.1038/sdata.2016.18 What does it mean “To be Reusable”: 1. (Meta)data is richly described with a plurality of relevant and accurate attributes (ontologies/controlled vocabularies) (e.g., metadata should describe the context under which the data was collected or generated), 2. (Meta)data is published with a accessible and clear data usage license (e.g., CC 0, CC BY 4.0, etc.), 3. (Meta)data is associated with detailed provenance. In simplest terms, “How easy is it to reuse the data?”. Research Data Life Cycle Research Data Life Cycle 18 RDM - Benefits and consequences? ¹ Research data lifecycle. https://libguides.ntu.edu.sg/rdm/researchdatalifecycle ² Bobrov, E., Adam, L.-S., Söring, S., Jäckel, D., Herwig, A., Lindstädt, B., Vandendorpe, J., & Shutsko, A. (2021). Workshop on Research Data. NFDI4Microbiota - Knowledge Base (https://knowledgebase.nfdi4microbiota.de/Research-Data-Management/02rdm.html#NTU_LibGuides_RD_life_cycle) Benefits and consequences of poor RDM 19 Data Management Plans - DMPs ¹ Assmann, C., Gadelha, L., Markus, K., & Vandendorpe, J. (2022). Workshop on Research Data Management. ² Bobrov, E., Adam, L.-S., Söring, S., Jäckel, D., Herwig, A., Lindstädt, B., Vandendorpe, J., & Shutsko, A. (2021). Workshop on Research Data. ³ Bres, E., Rudolf, D., Lindstädt, B., & Shutsko, A. (2022). Research Data Management in Medical and Biomedical Sciences. ⁴ Engelhardt, C., Biernacka, K., Coffey, A., Cornet, R., Danciu, A., Demchenko, Y., Downes, S., Erdmann, C., Garbuglia, F., Germer, K., Helbig, K., Hellström, M., Hettne, K., Hibbert, D., Jetten, M., Karimova, Y., Kryger Hansen, K., Kuusniemi, M. E., Letizia, V., … Zhou, B. (2022). D7.4 How to be FAIR with your data. A teaching and training handbook for higher education institutions. https://doi.org/10.5281/ZENODO.6674301 ⁵ Lindstädt, B., Vandendorpe, J., & von der Ropp, S. (2019). Research Data Management. ⁶ Jacob, B., Kroehling, M. A., Mertzen, D., Straka, J., Lindstädt, B., Shutsko, A., & Vandendorpe, J. (2022). Workshop on Research Data. ⁷ Voight, P., Frericks, S., Lindstädt, B., Shutsko, A., & Vandendorpe, J. (2022). Workshop on Research Data. Benefits (¹,²,³,⁴,⁵,⁶,⁷): -Visibility and clear data ownership -Data quality assurance -Eligibility for funding (legal incentive) -Prevents data loss -Saves time, money and resources Consequences: Data Management Plan(s) - DMPs 20 DMPs - What do they include? Required by DFG (since 2022) and EU Funding Programs (since 2021). They act as reporting tools for funding agencies, to hold grant recipients accountable to conduct good and open science. A companion for proposal writing to sharing of data and findings. DMP - Content 21 How to make a DMP? -Responsibilities and obligations (who, when, where) -Description of the research project (why, how, what) -Costs and resources (data generation, personnel, etc. ) -Description of the research data (type, quality, organization and usage) -Metadata to be collected -Storage and security (where will it be stored short term, who has access) -Digital preservation (long term storage/archiving, data preservation) -Legal aspects and anonymity (when dealing with sensitive data) ¹ Guides for Researchers. How to find a trustworthy repository for your data. https://www.openaire.eu/find-trustworthy-data-repository ² England, J., & Tsoukala, V. (2023). Horizon Europe Open Science requirements in practice - OpenAIRE webinar (Version 2023-11). Zenodo. https://doi.org/10.5281/zenodo.10125224 Generating a DMP 22 Data organization Research Data Management Organiser - https://rdmorganiser.github.io/en/ NFDI4MicrobiotaPlan - https://thoelken.github.io/dataplan/ → Exercise, make your own DMP using this tool. What is it missing. Example of a good DMP (or is it?): Molin, E. (2018). Behave Working Data-Management-Plan. Zenodo. https://doi.org/10.5281/ZENODO.1243717 DMP templates: 1. Biological & Environmental Sciences a. German Federation for Biological Data (GFBio): https://dmp.gfbio.org/ b. DataPlant: https://nfdi4plants.de/dataplan/ 2. Health Sciences a. University of Minnesota (incl. School of Public Health): https://www.lib.umn.edu/services/data/dmp-examples b. Clinical trials i. National Institutes of Health (NIH): https://www.nidcr.nih.gov/sites/default/files/2018-03/clinical-data-management-plan-template_0.docx ii. PAPA-ARTiS: https://ec.europa.eu/research/participants/documents/downloadPublic?documentIds=080166e5b6899b9b&appId=PPGMS Data organization - 5s methodology ¹⁽ ⁾ 23 File naming 1. Sort: delete unnecessary files. 2. Set in order: develop and document naming conventions and folder structures. 3. Shine: a. Comply with conventions. b. Develop routines. 4. Standardize: a. Document rules and responsibilities. b. Develop best practices and Standard Operating Procedures (SOPs). 5. Sustain: a. Regularly check whether rules are followed. b. Implement improvements if necessary. ¹ Lang, K., Roman, G., Jessica, R., Annett, S., Nadine, N., & Lehmann, A. (2021). The 5S Methodology in Research Data Management. Zenodo. https://doi.org/10.5281/zenodo.4494258 File naming 24 File name examples 1. Alphabetically sortable names are favourable (e.g. date YYYY-MM-DD) 2. Advisable to use up to 32 characters (e.g. 32CharactersLooksExactlyLikeThis.txt) 3. Use name that is unique to the content (e.g. 2024-07-09_sample_NODE_R16_A4.ffn) 4. Don’t use periods in the name, only in the extension 5. Special characters (“,|,&,%,$, etc.) and whitespaces (so space or tab) are confusing (for computers and others) 6. Use leading zeros (e.g. from 0001 - 0010 - 0100 - 1000) ¹ Assmann, C., Gadelha, L., Markus, K., & Vandendorpe, J. (2022). Workshop on Research Data Management. File name examples 25 Folder structure 1. Good structure ¹⁽ ⁾: a. YYYY-MM-DD_JV_ProjectID_ExperimentID with IDs being linked to a table with data documentation such as metadata 2. Good names ²⁽ ⁾: a. 2016-01-04_ProjectA_Ex1Test1_SmithE_v1-0.xlsx b. 2000_USNM_379221_01.tiff c. USNM_379221_01.tiff 3. Bad names ² :⁽ ⁾ a. Test data 2016.xlsx b. Meeting notes Jan 17 c. Notes Eric.txt d. Final FINAL last version.docx ¹ Bobrov, E., Adam, L.-S., Söring, S., Jäckel, D., Herwig, A., Lindstädt, B., Vandendorpe, J., & Shutsko, A. (2021). Workshop on Research Data. ² Bres, E., Rudolf, D., Lindstädt, B., & Shutsko, A. (2022). Research Data Management in Medical and Biomedical Sciences. Data license - data and results or software 32 Data and results - which license to choose. By default any creator of data, software, writing or any other content involving a sufficient amount of creativity is the copyright owner of that content without having to declare the copyright explicitly. Defining or using a suitable license for published content usually has the benefit of giving all parties legal certainty and understanding of permission to use. So it is about whether and how others can use your data. In sciences, two categories of licenses can applied to either software or data and results. Publishing figures and articles in journals, usually requires accepting the license agreement of the publisher and involves either a complete transfer of rights on your own work or picking an open access journal with acceptable permissive licenses. Data and Results - Creative Commons (CC) licenses 33 Software licenses. -CC-BY: Credit must be given to the creator. -CC BY-SA: Credit must be given to the creator. Adaptations must be shared under the same terms. -CC BY-NC: Credit must be given to the creator. Only noncommercial uses of the work are permitted. -CC BY-NC-SA: Credit must be given to the creator. Only noncommercial uses of the work are permitted. Adaptations must be shared under the same terms. -CC BY-ND: Credit must be given to the creator. No derivatives or adaptations of the work are permitted. -CC BY-NC-ND: Credit must be given to the creator. Only noncommercial uses of the work are permitted. No derivatives or adaptations of the work are permitted. -CC0: Public domain dedication. Software licenses 34 How to keep track of all you (meta)data. Exercise: Go to https://opensource.org/licenses Divide yourself in “x” groups of “y” participants Find out what is the difference (if any) between: -MIT -GPLv2 Think in terms of : -reuse (commercial and non-commercial) -Pipeline and workflow implementation -Adaptation of work (and under which license the adapted work must be published) Electronic notebooks (ELNs) 35 Take home message! Software meant to document experiments and research data. Some resources already available (NFDI4Microbiota in development) -SciNote ELN: A cloud-based ELN with lab inventory, compliance, and team management tools -Labguru ELN: An intuitive and user-friendly ELN with a focus on flexibility and adaptability -Labfolder ELN: A cloud-based ELN with a central position in daily research workflows, enabling easy sharing of information Take home message 36 Further reading CC-BY 2.0, https://www.flickr.com/people/33255628@N00 https://en.wikipedia.org/wiki/ File:Metadata_is_a_love_note_to_the_future_(8071729256).jpg Resources for further reading - links, explanations and more NFDI4Microbiota - Knowledge Base NFDI4Microbiota - MetadataStandards