Contributors: BIBI (lead, 62 PM), ZB MED (co-lead, 36 PM), UFZ (28 PM), FSU Jena (24 PM), JLU (11 PM), UMR (36 PM), EMBL (40 PM), RWTH (44 PM), HZI (30 PM), DSMZ (30 PM)

Goals: To coordinate, organize and document training events in NFDI4Microbiota to create a comprehensive training program. To promote online courses and online material, and to raise awareness of the value of the FAIR principles and appropriate strategies for RDM.

Training, i.e. the organized process of educating a target audience such that each individual is thereafter able to perform a particular task on their own, is a key means by which NFDI4Microbiota will connect with the scientific community, which in turn is the main objective of all NFDI efforts. Providing an extensive and broad scope of high-quality training is consequently one of the main measures of NFDI4Microbiota, as highlighted by the extensive contributions of each co-applicant. One of the first tasks will be the establishment of a single platform on which all training events provided by the network will be announced. This platform will be integrated into the central NFDI4Microbiota Hub (M3.1 - Central web portal) and, where possible, also link to comparable platforms of other NFDI initiatives. This platform will also hold documentation of past events, in particular by providing an archive of workshops, conferences, and training courses, including any materials available online. As training is a multiplier for building community trust in NFDI and for extending outreach, events will be actively promoted in collaboration with the WGCO (M1.3 - Community outreach and public relations) via e.g. the central Hub, social media and mailing lists. Our goal is to gradually reach out to specialist research centers (e.g. “Microbiome Signatures” SFB1371, “Gut-liver axis” SFB1382, or “Microverse” EXC2051), societies (e.g. the German Society for Hygiene and Microbiology (DGHM), the Association for General and Applied Microbiology (VAAM), the German Society for Virology (GfV)), and other consortia and universities. Once the announcement platform for past and future events is in place, we will work on an online training platform that enables organizers and experts to upload slideshows and online training materials, and to prepare full online courses on dedicated topics. This will allow researchers with different levels of knowledge to carry out their own off-schedule training and prepare themselves for courses with higher prerequisites.

The type of training events may vary, depending on the topic. They can be performed in person or virtually, in the form of workshops or summer schools. Some may include practical sessions such as hands-on lessons or hackathons. However, we will encourage organizers and presenters to provide online teaching materials as Open Educational Resources (OER) on hosting platforms like GitHub or ReadTheDocs and, once established, we will fully integrate these into the NFDI4Microbiota online training platform and archive them for the long-term. Additionally, and where practical, videos of the actual presentations will be recorded and added to the collection of online materials.

The training topics will cover different scopes and levels of detail, thereby addressing different user profiles, from students to group leaders, as well as different communities. The scope will range from general issues of research data management, data quality, bioinformatics and cloud accessibility to more NFDI4Microbiota-specific courses including e.g. microbial analyses, metagenomics and proteomics. The provision and management of FAIR data and the leverage of the FAIR principles, as well as the important aspect of experiment documentation and metadata standards, will be persistent and recurring themes throughout all training events.

In table 8 an overview of a potential first-year portfolio of NFDI4Microbiota training topics, covering specific training activities of which several are existing, upcoming events or are planned for the near future is provided. This overview also highlights the readiness of each co-applicant to substantially contribute to the overall training and education efforts with further resources. Consequently, each coapplicant will, on average, organize or significantly contribute to four specific training events per year, irrespective of whether these are in-person, virtual, or new online courses. In order to mitigate topic redundancy and temporal overlaps, the coordinator will actively support partners in creating and offering joint training events, consolidating the partners’ expertise to achieve the best training experience and knowledge transfer. Additionally, a large number of courses will be given in cooperation with other NFDI networks (e.g. NFDI4Biodiversity, NFDI4DataScience, NFDI4Health or GHGA) and institutes or research networks (e.g. DZIF), and coordinated closely with other national, European and international relevant training activities, e.g. through the de.NBI training platform and ELIXIR TeSS. Table 8

NFDI4Microbiota will also coordinate a mentorship program, where mentors will advise mentees on scientific and career matters. Mentees will be paired with mentors of the appropriate field and career stage from different institutes. Mentors and mentees will be recruited directly from participant and coapplicant institutes, as well as through attendance at training courses and via a sign-up system on the online web portal (M3.1 - Central web portal). Virtual training courses will guide mentors and mentees on how they can each contribute and gain the most from the program. A suggested schedule of virtual meetings will be provided, as well as suggested topics for discussion. The mentorship program will be open to all, but extra effort will be made to include and support individuals from underrepresented groups in science. This program will enhance connectivity throughout the NFDI4Microbiota community and support one-on-one training.

To guarantee high-quality training and education, we will strive to implement three additional measures. First, we will develop general good practice guidelines for training, most probably in the form of a white paper. This will help training organizers to set up events and help contributors to use tried-and-tested solutions for their presentations and FAIR data sharing. Second, together with M1.3 - Community outreach and public relations and M4.4 - Service monitoring & reporting, we will establish a training feedback form to gather feedback in a consistent and comparable manner (analogous to the de.NBI training surveys). This will allow us to respond directly to feedback, report back to organizers, and check the overall quality of the training based on user assessments. Third, we will work closely with The Carpentries, an international not-for-profit organization that teaches foundational coding and data science skills to researchers and librarians worldwide *27. It offers a detailed, evidence-based teaching methodology, including programs to train instructors and maintain OER teaching materials. Members of NFDI4Microbiota are heavily involved in this community and the consortium will build upon these connections. Moreover, we will collaborate with ELIXIR throughout the “Train the Trainer program” (TtT) and plan to establish at least one annual event to communicate the good practice guidelines more directly. This also has the potential to serve as a certificate for NFDI4Microbiota trainers, providing credits toward scientists’ careers. Furthermore, we will collaborate with Open Science MOOC *28 to work on common content and formats. We will also invite external teaching experts to evaluate the training program and expand our content and methodologies based on their feedback. Table 9

Contributors: ZB MED (lead, 30 PM), DSMZ (co-lead, 30 PM), UMR (12 PM), all other co-applicants (sig. in-kind)

Goals: To offer a broad spectrum of support to overcome individual problems faced by researchers including the writing of data management plans, data submission, metadata collection, and legal issues (e.g. data licenses). To help with the FAIRification of databases and provision of dedicated software tools.

Operations: The support team will consist of a help desk maintained by data stewards who can assist directly with the majority of requests. The help desk will use a ticket system to forward requests to the partner that is most qualified to offer support in the respective area, e.g. the specialists in training and education (M1.1 - Training and education) or in services (Task area 3 - Services). NFDI4Microbiota will also reach out to other NFDI consortia to enable requests to be forwarded to them if they have the required expertise and, vice versa, to receive requests from their user community.

Online community: A help forum will be established as part of the NFDI4Microbiota Hub (M3.1 - Central web portal) that will enable users to support and interact with each other, solve common issues on their own, and enliven the community with discussions on the most suitable strategies for solving common challenges. Users can post issues of specific or general interest, which can be answered either by other users or NFDI4Microbiota moderators. Over time, this will lead to the development of an extensive knowledge resource. NFDI4Microbiota ambassadors (M1.1 - Training and education) will carry this multiplier effect into their local institutions and foster use of the forum. Private issues can be directed to our central, multi-level help desk and will be tracked via an internal ticketing system (MS1.2.1). To enable help desk agents to thoroughly answer user questions, we will compile comprehensive, welldocumented catalogs of available analytical services – both those hosted by consortium members and others – together with collections of SOPs (M2.2 - Experimental procedure standards) and best practices (as e.g. benchmarked in M3.2 - Analytical services). Agents might point users to suitable upcoming training events (through our central calendar), online tutorials and matching forum threads or suggest use of electronic lab notebooks hosted by NFDI4Microbiota.

Data and metadata management: In order to ensure the high quality of data and metadata, NFDI4Microbiota will support participating institutions and other members of the community at all stages of their research. This also includes the structuring of metadata and referrals to further experts, if needed. Especially in the projects that will be performed together with the participants (M1.7 - Use cases), the data stewards will be tightly embedded in data management, helping to generate data management plans, providing advice on metadata collection and helping with the actual data and code deposition.

Dataset/database/software FAIRification: The support team will also assist researchers during all the phases of a database’s life cycle, focusing on FAIRification, safeguarding of data, research software and knowledge (M3.3 - Databases & terminology services). Established databases can be further improved e.g. by search machine optimization, standardization of data, extensive linking, and web service development. Integration of databases into the central web portal (M3.1 - Central web portal) will be encouraged and actively supported. Old datasets that are hard to reuse due to technical or legal issues can also be rescued and made FAIR, either by generating simple machine-readable files or by including the data on Wikidata *29 or on the Open Research Knowledge Graph (ORKG) *30. ZB MED and TIB have started a Microbiology Observatory for this purpose in ORKG.

Evaluation: The support activities will also be regularly evaluated to develop new services (Task area 1 - Community, Networking & Training), standards, training materials (M1.1 - Training and education), or other activitiesThe support activities will also be regularly evaluated to develop new services (Task area 1 - Community, Networking & Training), standards, training materials (M1.1 - Training and education), or other activities. Table 10

Contributors: HZI (lead, 30 PM), ZB MED (co-lead, 24 PM), FSU Jena (7 PM), all other co-applicants (sig. in-kind)

Goals: To engage a diverse target community, provide information on NFDI4Microbiota’s mission and highlight the advantages of FAIR and standardized data sharing for data holders and platform users. Active community engagement will ensure that platform functionalities and services are in line with user expectations and needs, thus promoting active use and community contributions. To this end, NFDI4Microbiota will establish a Working Group for Community Outreach (WGCO), responsible for organizing and overseeing community outreach and engagement and promoting consortium activities.

The WGCO will include the lead and co-lead of this measure, as well as the lead for Service Monitoring & Reporting (M4.4 - Service monitoring & reporting). Its tasks will be i) to inform and promote NFDI4Microbiota activities via various channels, ii) to coordinate community engagement via workshops, surveys, and training, and iii) to gather and summarize feedback from platform users, such that it can be used as part of efforts to improve the platform by the technical task areas responsible for data storage and service implementation (M4.1 - Computational infrastructure operations, M4.2 - Data storage platform, M4.3 - Infrastructure software components).

Informing, promoting and gathering feedback: NFDI4Microbiota already operates a general website *26, offering information on services to engage the community and providing state-of-the-art information on microbiome-related data and their analysis. This site will be expanded into a central portal that will act as the point of entry to all other NFDI4Microbiota resources and services, such as data upload and download, cloud-based analyses and workflows, and training courses (M3.1 - Central web portal). The site will additionally feature informative videos and regular podcasts on microbiomerelated topics. Furthermore, NFDI4Microbiota will promote its activities to different user communities on social media (e.g. Twitter), through its established mailing list, in press releases, and via other channels. Four community workshops have already been hosted to communicate and discuss NFDI4Microbiota’s mission. The feedback from these events and further questionnaires has been incorporated into this proposal. NFDI4Microbiota will continue to implement community feedback from various channels, such as the USIGs, the SAB and further questionnaires.

NFDI4Microbiota Ambassador Program: NFDI4Microbiota will establish the NFDI4Microbiota Ambassador Program to create a direct connection to participant institutions. The aim is to engage young researchers (PhD students or postdoctoral researchers) as local contacts at participating institutes and other institutes interested in the program. Local ambassadors will get dedicated training in the core topics relevant to research data management and the services provided by NFDI4Microbiota. They can communicate what they learn in local internal meetings, colloquia and individual meetings. They also represent an additional point of contact with NFDI4Microbiota users. For the institutions, ambassadors can serve as experts to consult on NFDI4Microbiota topics. All participating institutions have agreed to provide at least one person to act as an ambassador.

Data providers: One of the main NFDI4Microbiota goals is to promote synergistic and responsible data sharing and analysis, and to clarify the value of these processes for data holders and analysts. To foster acceptance among data generators, it is important to guarantee and also communicate that data owners will keep control over their data and get to decide when their data will be released to the public. We will also closely interact with key data contributors such as the Next Generation Sequencing Competence Network (NGS-CN) to ensure that NFDI4Microbiota accepts and complies with their wishes and needs. Together with the different user communities, we will identify the most relevant datasets that should be represented in the platform, as well as required analysis features (M3.2 - Analytical services).

Engaging the communities: To promote and strengthen the consortium and its goals, several events with different objectives are planned. An annual international meeting will gather NFDI4Microbiota partners with the diverse user communities, presenting an opportunity for direct interaction between consortium members and other scientists. The SAB will be part of this annual meeting to strengthen its ties to the community, so that it is aware of and can thus advise on possible concerns and wishes.

Microbiology-related communities: NFDI4Microbiota’s target audience includes the diverse set of microbiology-related and application-related communities of e.g. bacteriologists, virologists, clinician scientists, epidemiologists, parasitologists, mycologists, and microbial ecologists who focus on clinical, environmental, agricultural and biotechnological applications. Among these users are the key data contributors and expert users who need to be informed of NFDI4Microbiota use cases and encouraged to deposit data and use the analytical services, as well as give feedback. To reach the individual groups, the consortium will be present at important community conferences, such as the Annual Meeting of the German Society for Hygiene and Microbiology (DGHM) and Association for General and Applied Microbiology (VAAM), the Annual Meeting of the Society for Virology (GfV), the International Virus Bioinformatics Meeting (EVBC), the Scientific Conference of the Mycological Society (DGfM), the DGEpi-Online-Jahrestagung for epidemiologists and the Annual Meeting of the German Society for Parasitology (DGP), which will be extended to include e.g. events focusing on biotechnology, agriculture, clinical research, environment and ecology. Here, posters will be presented or an information booth provided to promote NFDI4Microbiota topics and enable direct contact with consortium representatives, and discussion of needs. Links to existing consortia, community activities, such as participating SFBs, SPPs and EXCs will be leveraged. The different societies will also be addressed individually at the annual NFDI4Microbiota event, with contributions and workshops specific to their application scenarios. Further communication of the NFDI4Microbiota mission and goals will be achieved through societyspecific newsletters and mailing lists, as well as social media posts. This measure will be connected to M4.4 - Service monitoring & reporting, ensuring that user wishes are communicated and rapidly implemented into the platform where applicable.

Biobanking community: Jörg Overmann from DSMZ will leverage his contacts with the biobanking community and promote NFDI4Microbiota’s goals on FAIR data sharing, as well as active use of the platform. We will explore synergies with the different biobanks, such as linking biobanked material and metadata with corresponding data deposited via the NFDI4Microbiota infrastructure.

Data science, bioinformatics and computational biology communities: This target audience, similar to the diverse audience of microbiologists, will be informed of NFDI4Microbiota use cases and encouraged to deposit data and use the analytical services, as well as to provide feedback as representatives of the technical side of microbiome research. The bioinformatics target audience will also be addressed to identify or further add to the core services offered on the NFDI4Microbiota platform.

Engaging citizen scientists: To include a more general audience, the NFDI4Microbiota Hub (M3.1 - Central web portal) page will include contributions designed for a more general public, featuring informative videos on the consortium and user surveys. Further outreach will be realized via press releases, newsletters, blog posts, videos and podcasts, and events such as a microbiology science slam.

Harmonizing multiple communities: To establish a successful and harmonized NFDI4Microbiota infrastructure used by various target communities, their requirements need to be unified as far as feasible and sensible. To achieve this, we have defined, and will further extend, use cases (M1.7 - Use cases) of interest to multiple communities. NFDI4Microbiota will closely collaborate and work in line with existing European and international organizations, such as ELIXIR, EOSC-life and other NFDI consortia (M1.4 - Connection to other NFDI consortia) to harmonize data access on a global and long-term basis. Table 11

Contributors: HZI (lead, 27 PM), ZB MED (co-lead, 6 PM), FSU Jena (6 PM), UMR (3 PM), JLU (2 PM), BIBI (4 PM), EMBL (3 PM), UFZ (3 PM), DSMZ (sig. in-kind)

Goals: To make NFDI an effective and sustainable endeavor, the different consortia need to closely collaborate and work in a trans-disciplinary manner. NFDI4Microbiota will closely collaborate with others, align activities, and interconnect services (The consortium within the NFDI).

Cross-cutting topics: As a signatory to the Leipzig/Berlin declaration on NFDI cross-cutting topics (“Leipzig-Berlin-Erklärung zu NFDI-Querschnittsthemen der Infrastrukturentwicklung”) NFDI4Microbiota will actively join the discussion and implement shared topics. We will contribute to common solutions by sending representatives to inter-consortia conferences, workshops and working groups, and assist in organizing such events. The outcome of these discussions will be included in our activities.

Bilateral/multilateral collaboration with other consortia: We will explore synergies and harmonize activities with other consortia on all levels, e.g. training activities, achieving compliance, and further extending existing data and workflow standards. This will be facilitated by the fact that several NFDI4Microbiota co-applicants are also active within other NFDI consortia and joint activities are already planned. Implementing joint use cases (M1.7 - Use cases) serves to generate common solutions and achieve interoperability between solutions from different consortia. We will continuously work on new use cases that help develop common standards as well as their technical implementations. We will also work with GHGA, DataPLANT, NFDI4Biodiversity, and the international community to further develop training portfolios as well as metadata and workflow standards.

Synergies with NAKO, GHGA and NFDI4Health: To further leverage existing ties between NFDI and the epidemiology community, we will work together with NFDI4Health and NFDI4Medicine. NFDI4Microbiota will work with the NAKO (Nationale Kohorte) health study, which aims to investigate the causes of chronic diseases through extensive sampling of the German population and by generating different types of omics data, including microbiota-related data types. GHGA is already collaborating with NAKO in the analysis of human omics data and NFDI4Microbiota will explore synergies with both consortia.

To this end, we will generate and promote informative material on FAIR data sharing and the advantages this offers, thereby responding to and alleviating the fears and obstacles currently hindering rapid and FAIR data deposition practices. One of the main aims here is to ensure that future data deposition will occur in data repositories that follow FAIR principles and standards established by NFDI4Microbiota and other NFDI consortia. Table 12

Contributors: FSU Jena (lead, 30 PM), UFZ (co-lead, 17 PM), UMR (3 PM), JLU (2 PM), RWTH (in-kind)

Goals: To connect NFDI4Microbiota with the international scientific landscape, maximize synergies and determine how to best integrate efforts across all areas, e.g. standards, analytical workflows, data and result sharing, as well as training efforts.

We will collect feedback to define key concepts in the various measures, such as M2.1 - Data & metadata standards, M3.7 - Reviewing/commenting system for data, M3.3 - Databases & terminology services, M2.2 - Experimental procedure standards. In addition, training courses offered by NFDI4Microbiota will be devised in collaboration with international experts (M1.1 - Training and education).

Working groups: Leading national and international scientists will be invited to engage with the NFDI4Microbiota network to discuss emerging topics, foster creativity, and identify cutting-edge trends in microbiota research. They will also receive support for their own analyses from NFDI4Microbiota. The working groups will meet six times in the first year, three times in the second year, and then once a year after that. This timeframe can be adjusted to reflect emerging topics during the funding period.

Example 1: To develop a database for bacteriophages, we need to develop a new data structure. We make agreements with NCBI, EMBL/EBI, and ViPR database developers to discuss this in detail in a three-day workshop in month 6.

Example 2: Human microbiome research advances at an unprecedented pace. To keep up with the cutting-edge science in the field, we create a working group linked to the latest developments in standard operating procedures designed to optimize data quality and analytical workflows for processing human microbiomes, thereby increasing comparability. HZI, EMBL and UFZ (all co-applicants of NFDIMicrobiota) would be involved in such a working group, together with the ELIXIR metagenomics community and international NFDI4Microbiota partners, such as Curtis Huttenhower (Harvard), Rob Finn (EMBL-EBI), Nikos Kyrpides (JGI), Niranjan Nagarajan (A Star Genomics Institute Singapore), the BGI, and Rob Knight (UCSD).

Dedicated events: We will organize a five-day kick-off meeting (one week, partially remotely) including international partners as keynote speakers. It will be attended by all NFDI4Microbiota co-applicants and participants and other target-community scientists who express interest. This event will be combined with the first meetings of the different working groups. Subsequently, we will organize international conferences in Germany annually, with international partners and other key experts. Table 13

Contributors: HZI (lead, 12 PM), EMBL (co-lead, 10 PM), FSU Jena (3 PM)

Goals: To develop and implement strategies for establishing NFDI4Microbiota as a long-term hub for microbial research in Germany.

All infrastructure and service projects face the challenge of how to continue operations after the original funding phase. To enable NFDI4Microbiota services to be offered to the community over the long term, NFDI4Microbiota will study different ways of applying them. The modular and service-oriented nature of the planned NFDI4Microbiota infrastructure will facilitate long-term maintenance, and thus sustainability. We will investigate the acquisition of funding from the federal government and/or states for the entire infrastructure and for individual components, such as the data storage infrastructure or analysis toolbox. We will also explore common solutions with other NFDI consortia. To ensure visibility, we will promote NFDI4Microbiota’s mission and progress to decision-makers and funding bodies, such as the Federal Ministry of Education and Research (BMBF) and the German Council for Scientific Information Infrastructures (RfII) *31, which focuses on the topics “Research Data - Sustainability - Internationality” and represents a broad range of scientific disciplines, actors and institutions. Among other topics, this council provides recommendations on development of the NFDI, publishing discussion papers and giving recommendations on establishing successful NFDI consortia *32. The council has the role of a consultant on policy and science, but can also initiate discussion processes. To update stakeholders and decision-makers on NFDI4Microbiota proceedings, we will provide annual reports including challenges faced and progress made. In the planned community activities (M1.3 - Community outreach and public relations), i.e. community workshops, and the annual international meeting, we will gather representatives from important communities, such as microbiology, virology, mycology and medical, agricultural and biotechnological societies, but also from more distant fields that could make use of the NFDI4Microbiota infrastructure and community interactions. Through these means, we aim to convey NFDI4Microbiota challenges and progress and maintain an open discussion with a range of communities to establish a widely used platform over the long term. Table 14

Contributors: RWTH (lead, 60 PM), ZB MED (co-lead, 15 PM), all co-applicants (6 PM each)

Goals: To employ all the measures offered by NFDI4Microbiota to support cutting-edge scientific projects (use cases) that involve a wide array of scientists from multiple disciplines.

This measure will help put the solutions proposed by NFDI4Microbiota into practice. All the measures converge here to serve the purpose of various use cases that represent a broad range of projects, all of which require the support of microbiota data infrastructures and services. Each use case addresses an issue of major scientific and/or global relevance that is best tackled by combining the expertise of several stakeholders acting as a small consortium with a designated lead. In all cases, NFDI4Microbiota will provide training (M1.1 - Training and education) as well as support and consulting opportunities (M1.2 - Support) to all those who need them within the community. Beyond this general support, depending on the designated goal and anticipated challenges (as mentioned in the detailed descriptions) each use case calls for different NFDI4Microbiota measures that will help overcome the main project challenges in the best possible manner.

The use cases bring together all stakeholders involved in NFDI4Microbiota, including co-applicants, participants, national research consortia (e.g. collaborative research centers, priority programs, clusters of excellence), and societies within the broad field of microbiology. Importantly, they will act as catalysts for fostering important interactions with several other NFDI consortia, in which the handling of microbiota-derived data is highly relevant but not specifically tackled.

This use case measure will not only allow the community to benefit from all the other measures implemented by NFDI4Microbiota, but will also foster improvements of these measures thanks to community feedback. Eventually, this will lead to fine-tuning of the measures over time to best address the community needs via concrete examples of scientific work performed in many labs across the country. Hence, the use cases represent a core activity within NFDI4Microbiota, for which efficient coordination is key. A lead institution has therefore been assigned to each use case, corresponding in most cases to co-applicants (since they know the consortium best and thus understand how to implement measures and coordinate actions most efficiently), but also sometimes to participants who can inject very specific expertise and carry out projects of relevance to many. This use case-specific leadership will take place under the umbrella of the measure lead at RWTH who will drive long-term commitment and efficient communication between the partners.

To implement a dynamic system that stays on track with new trends in the field, this measure will integrate new use cases from the community as they arise over time, either directly through NFDI4Microbiota activities or independently. To offer a platform for such new use cases, yearly calls for applications will be issued from year 2 onward (i.e. year 2, 3, and 4). Successful use cases will be selected via a review process organized by the board of co-applicants and will receive financial support for a duration of 1 to 2 years via the flexible funds available within NFDI4Microbiota. Such a system will guarantee flexibility and give the opportunity to broaden the scope of activities to an even wider pool of participants, increasing the visibility and, eventually, the added-value of the consortium. Table 15

5.1.7.1 Use case ’AdaSPat’: Phylodynamics of viral pathogens in clinical isolates

Scientific relevance: NFDI4Microbiota enables comprehensive and standardized genome-based analyses of pathogens, supporting discovery of treatment- and pathogenicity-related polymorphisms.

Case: Lilian is a clinician who has sequenced genomes of human cytomegalovirus from patient samples. She configures a tailor-made NFDI4Microbiota workflow to perform quality controls, remove human data, which falls under the GDPR, and perform viral haplotype assembly to identify mixed-strain infections. She then identifies several polymorphisms associated with changes in pathogenicity, virulence or resistances using a screen against a mutation catalog. Furthermore, the data is used together with public data to indicate putative transmission clusters.

Linked institutions and network: HZI (lead); FSU Jena; JLU; GHGA; UKJ; NFDI4Health.

Challenges and NFDI4Microbiota measures:

• Implement configurable data analysis workflows (M2.1; M2.3; M3.4).

• Interface for data visualization (M3.1; M3.2; M4.1; M4.3; M3.5).

• Manage data falling under GDPR (M2.1; M2.4; M3.5; M3.6; M4.2).

5.1.7.2 Use case ‘ATTRACTOR’: Large-scale integration of metagenomes

Scientific relevance: The number of individual microbial studies based on sequencing technologies is multiplying at a rapidly increasing rate. The research community has not yet fully benefited from this unprecedented amount of sequencing knowledge because of the lack of approaches for efficient integration and easy exploration of the data.

Case: Ilias is a young bioinformatic group leader interested in studying global diversity within microbiomes, including prokaryotes, eukaryotes, and viruses via large-scale integration of metagenomic data. In particular, he wants to prove the existence of persistent composition states (attractors) in different environments, determine the core taxa therein, and study their functional interactions. An existing public tool created during his postdoc provides a solid foundation for this project *33. However, he now faces the multiple challenges listed below. NFDI4Microbiota will support his work at this early stage of his academic career while also benefiting a wide variety of potential users within the community.

Linked Institutions & network: TUM (lead); RWTH; FSU Jena; EMBL; ZB-MED; SFB1371; SFB1382; SFB1076 (via FSU Jena); DataPLANT; NFDI4Earth; NFDI4Agri; NFDI4Biodiversity; VAAM.

Challenges and NFDI4Microbiota measures:

• Search, access, and curate publicly available datasets and corresponding metadata (M2.1; M2.4; M3.4).
• Rapid processing of all collected metagenomics data according to a standardized workflow (M2.3; M3.2;M3.3).
• Expand the existing interface by incorporating novel features for analyses and visualizations (M3.5;M3.8; M4.3; M3.5).
• Guarantee reliable long-term hosting of the platform (M3.6; M4.1; M4.2).

5.1.7.3 Use case ‘BIOCAT’: Biocatalysis

Scientific relevance: Novel enzymes are of central importance for making progress in biocatalysis. As well as using protein engineering methods to improve enzymes, scientists are also seeking to harness a plethora of as-yet undiscovered biocatalysts from microbes.

Case: Arthur uses high-throughput technologies to characterize novel enzymes in the Bornscheuer lab, a partner in the NFDI4Cat consortium, which already receives funding. NFDI4Microbiota will support him in studying enzymes involved in flavonoid biosynthesis such as chalcone isomerases, for which only one enzyme has been fully characterized to date. His work generates a vast amount of data that require management. Moreover, a preliminary phylogenetic analysis identified >500 candidate proteins in need of further exploration.

Linked institutions & network: Uni Greifswald (lead); RWTH; FZ Jülich; NFDI4Cat; VAAM.

Challenges and NFDI4Microbiota measures:

• Identify candidate genes/proteins from genomic and metagenomic data (M3.4; M3.7; M4.1).
• Build sophisticated phylogenetic trees (M3.2; M3.3).
• Study candidate proteins, link to biochemical and biocatalytic features, update databases (M2.1; M3.6).

5.1.7.4 Use case ‘CLINIMIC’: Clinical microbiomes

Scientific relevance: The number of clinical samples collected for microbial community analyses is increasing rapidly. Hence, it is essential to establish standardized operating procedures (from biobanking to data analyses) to study microbial diversity and characterize important functional traits such as antimicrobial resistances and other pathogenicity factors linked to specific clinical conditions.

Case: The University Hospital Gießen/Marburg coordinates the establishment of a nationwide biobank of bronchoalveolar lavage fluids (BALF) and paired blood samples from patients with pneumonia to identify microbial pathogens of the lower respiratory tract in combination with immune cell phenotyping by cytometry. To gain comprehensive insights into microbial communities and the virome in these samples, microbiome sequencing with different SOPs needs to be performed.

Linked institutions & network: EMBL (lead); JLU; BIBI; RWTH; UMR; FSU Jena; UK Frankfurt; UK Erlangen; UK Köln; HKI; MDC Berlin; SFB1371; SFB1382; EXC2051; GHGA; NFDI4Immuno; NFDI4Health; DGHM.

Challenges and NFDI4Microbiota measures:

• Optimize sample collection and metadata acquisition (M2.1; M2.3).
• Generate standardized microbiome profiles (M3.2; M3.5; M4.1).
• Compare with thousands of other data sets from public repositories based on securely stored clinical data (M2.4; M3.1; M3.3; M3.4; M3.8).

5.1.7.5 Use case ‘CRISPR’: From sequence-based diversity to microbial engineering

Scientific relevance: Microbiome research is very much hampered by the fact that only a few standard bacterial species exist as models. The vast majority of unknown commensals present in native communities represent a yet unexplored pool of functions. When cultured, commensals are often difficult to work with and manipulate, which complicates the performance of mechanistic studies.

Case: Melissa is a microbiologist who specializes in isolating and studying gut bacteria from different animal hosts by combining (meta)genomics and cultivation. Besides working on genome-based taxonomy of all her isolates and studying the sequence-based diversity of CRISPR-Cas sequences to unravel novelfunctions beyond defense, she has also identified new genes encoding lipolytic enzymes. She now wants to knock these out via phage-based delivery of target CRISPR-Cas systems and investigate their expression via RiboSeq and quantitative proteomics, prior to studying effects on the host in gnotobiotic mice colonized with a synthetic community.

Linked institutions & network: RWTH (lead); DSMZ; UMR; JMU; FZ Jülich; SPP2330; SPP2141; NFDI4Biodiversity; VAAM.

Challenges and NFDI4Microbiota measures:

• Curation and management of generated genomes (M2.1; M2.4; M3.5; M3.6; M4.2).
• Detailed analysis of target genes (M2.3; M3.2; M3.3; M4.1).
• Development and use of molecular engineering systems (M2.2; M3.9).

5.1.7.6 Use case ‘DataSci’: Data Scientist

Scientific relevance: Analysis of high-throughput data obtained from various technologies requires a multitude of bioinformatic tools and databases. The NFDI4Microbiota platform will support data scientists in their daily work by providing standardized computing environments and by facilitating data management and workflow execution, regardless of the specific domain, with scalable, high-performance storage and an execution backend.

Case: Mark, an experimental biologist, has no experience with omics data analyses. He thus collaborates with Julia, a data scientist. Together they use the NFDI4Microbiota platform, which helps them manage the data and execute workflows. The platform offers a way for Mark to upload his data via a GUI or CLI, including updates and versioning, to generate reproducible results. The uploaded data is automatically analyzed via execution of a workflow created by Julia within the NFDI4Microbiota infrastructure. Julia also implemented a small web application to visualize results and deployed it via a Docker image. The versioning of all input data makes it possible to display the most recent data and compare all the data over time.

Linked institutions & network: JLU (lead); BIBI; EMBL; UK Essen; HHU; HIRI; JMU; FLI; LMU; MLU; MDC Berlin; MPI Magdeburg; TU Darmstadt; Uni Göttingen.

Challenges and NFDI4Microbiota measures:

• Train people to use the platform (M1.1).
• Build a scalable storage and computing platform with a user friendly WebUI; provide a deployment infrastructure including an authentication service (M3.1; M4.2; M4.3; M3.5).

5.1.7.7 Use case ‘GUT’: Gut microbiomes

Scientific relevance: As a substantial fraction of gut microbiomes is still uncharacterized, there is a dire need to implement innovative approaches to study the wealth of microbial diversity detected by shotgun metagenomics at the strain level.

Case: Jens, a microbiologist from Aachen, aims to explore the thousands of as yet unknown microbial species in the mammalian gut in order to study the host-specificity of intestinal microbiomes at the strain level. As illustrated by the multiple challenges listed below, reaching this goal will require him to master diverse wet lab and in silico skills and process a vast amount of data.

Linked institutions & network: RWTH (lead); DSMZ; BIBI; FSU Jena; EMBL; UMR; TUM; UK Essen; HHU; UK Erlangen; UK Frankfurt; JLU; CAU Kiel; HKI; MPI Magdeburg; UK Köln; IKMB; MDC Berlin; SFB1371; SFB1382; SPP2002; EXC2051; DGHM; NFDI4Biodiversity. NFDI4Health.

Challenges and NFDI4Microbiota measures:

• Generate metagenomes and high-quality metadata from hundreds of gut samples from different host species (M2.1; M2.2; M2.4; M3.2; M3.5).
• Build metagenome-assembled genomes (MAGs) from these data and thousands of other datasets from public repositories at a strain-level resolution (M2.3; M3.1; M3.2; M3.4; M4.1).
• Establish collections of well-curated genomes, taxonomically describe and validate novel bacteria, and compare genome and MAG catalogs (M3.2; M3.3; M3.6; M4.2).

5.1.7.8 Use case ‘MetaBench’: Benchmarking meta-omics software

Scientific relevance: The plethora of tools available for metagenomics generates confusion and redundancy and enhances the risk of skewed analyses due to the lack of standards. NFDI4Microbiota will provide a platform for comparing workflows based on evaluation packages and benchmarked datasets created via the CAMI initiative (Critical Assessment of Metagenome Interpretation). This will help developers to create best-practice guidelines for common analyses and easily benchmark new techniques.

Case: Carol is a computational biologist in Braunschweig working on improving a metagenome binner developed by her lab. The benchmarking framework for meta-omics software provided by NFDI4Microbiota will help her by allowing both comprehensive testing of the binner with standard datasets and comparison to existing approaches. If Carol’s method is rated one of the best for this kind of data, it will be further analyzed internally by the platform team and eventually included in the NFDI4Microbiota data processing workflows.

Linked Institutions & Network: HZI (lead); BIBI; RWTH; EMBL; MPI Magdeburg; TUM; SFB1371; SFB1382; VAAM.

Challenges and NFDI4Microbiota measures:

• Implement the benchmarking interface (M3.1; M4.1).
• Generate further benchmark data sets (M4.3; M3.5).
• Create benchmarking workflows (M2.3).
• Derive best practices for data processing (M4.4).

5.1.7.9 Use case ‘METAMAR’: Global impact of climate change on marine biogeochemical cycles

Scientific relevance: Current climate change models indicate that a number of marine biomes have been transformed beyond repair. A catalog of worldwide marine microbial life (prokaryotes, eukaryotes and viruses) will not only provide a basis for comparison with future climate scenarios, but also deliver conceptual knowledge on how changes in climate may alter the functioning of biogeochemical cycles on a global scale.

Case: Jose, a marine microbiologist from Bremen, is studying the functional and phylogenomic diversity of microbes in the ocean at a global scale. He therefore decides to explore approx. 16,000 marine metagenomes publicly available in ENA. Due to the importance of multi-trophic interactions and networks in a biogeochemical context, Jose opts to examine these datasets in a multi-domain manner by studying the genetic potential and phylogenomics of prokaryotes, eukaryotes and viruses. Jose expects to create a database of >75,000 novel genomes. He will use this data to explore the biogeography of microbes and viruses, and to hypothesize potential changes of biogeochemical functioning in the marine environment.

Linked institutions and network: UFZ (lead); Uni Hamburg; FSU Jena; NFDI4Earth; NFDI4Biodiversity.

Challenges and NFDI4Microbiota measures:

• Screen public repositories for soil metagenomes and standardized metadata (M2.1; M3.1; M3.4; M3.5; M3.7).
• Long-term storage of raw data (M2.1; M4.2).
• Simultaneous recovery of prokaryotic, eukaryotic and viral metagenome-assembled genomes (MAGs) followed by phylogenomics and functional annotation (M3.2; M3.3; M4.1; M4.3; M4.4).
• Archive the novel data and associated metadata in public repositories (M3.5).

5.1.7.10 Use case ‘METASOIL’: Mechanisms of antimicrobial resistance and degradation of pollutants in soils

Scientific relevance: Soil microbes are the primary drivers of organic material decomposition, efflux of CO2 and other greenhouse gases, and global nutrient cycling. A better understanding of the mechanisms controlling assemblage of soil microbial communities will provide the conceptual tools necessary to understand how anthropogenic influences affect relevant microbial functions for ecosystem health (such as antimicrobial resistance and degradation of pollutants).

Case: Heike, a microbial ecologist from Leipzig, is working on the mechanisms underlying community assembly in soils systems. She therefore decides to explore approx. 8,000 datasets of soil metagenomes publicly available in ENA. Using NFDI4Microbiota metadata standardization, she will classify these different soils according to different levels of antrophogenic influences. She will then examine these communities in a multi-domain context by studying the genetic potential and phylogenomics of prokaryotes, eukaryotes and viruses. To create a database with >50,000 novel genomes, Heike uses NFDI4Microbiota analytical tools to (i) accurately annotate the genetic potential underlying antimicrobial resistances, and (ii) define the potential multi-domain interactions involved in the metabolism of selected pollutants in soil systems.

Linked institutions and network: UFZ (lead); ZB MED; BIBI; EMBL; DataPLANT; NFDI4Earth; NFDI4Biodiversity; NFDI4Chem; NFDI4Ing; NFDI4Agri; VAAM.

Challenges and NFDI4Microbiota measures:

• Large-scale analysis of public soil metagenomes (M2.1; M3.1; M3.4; M3.7).
• Multi-domain phylogenomics and functional annotation (M3.2; M3.3; M4.1; M4.3).
• Submit novel data and associated to public repositories (M3.5; M4.2).

5.1.7.11 Use case ‘MIC-HOST’: Integration of host genetics and microbiome data

Scientific relevance: Connecting human genetics with microbiome research is scientifically powerful yet challenging due to the multiple fields of expertise and ethical considerations required.

Case: Supriya is a postdoc at University Hospital München Rechts der Isar. She is currently supervising a clinical study with 500 Crohn’s disease patients that aims to dissect the link between human genetics and gut microbial genes in inflammation. She has collected stool and blood samples and sent them for sequencing to the DFG-funded competence center in Tübingen. Patients have granted ethical approval for their data to be stored, analyzed, and integrated. She has collaborators who are experts in human genetics, Adaptive Immune Receptor Repertoires (AIRR), and microbiome analysis; Supriya herself will bring this expertise together and integrate the data.

Linked institutions: EMBL (lead); BIBI; RWTH; FSU Jena; HZI; MPI Magdeburg; TUM; UK Erlangen; Uni Hamburg; SFB1371; SFB1382; Frauenhofer CIMD; NFDI4Health; NFDI4Immuno; GHGA; DGHM.

Challenges:

• Upload patient clinical data and sample metadata in a secure and QM-compliant manner (M2.1; M3.4; M4.2).
• Integrate clinical, host genetic, and microbiome data of different formats and standards (M3.2; M3.8; M4.1).
• Synergize interactions & enable rapid data access and visualization by project partners (M3.1).
• Ensure future use of the data in compliance with safety and privacy guidelines (M2.4; M3.6).

5.1.7.12 Use case ‘MICIMG’: Microbial Imaging

Scientific relevance: Wastewater is contaminated by human activities. It contains physical and chemical pollutants and is discharged into rivers, lakes, and oceans that harbor important microbial communities. The effects of pollutants on these communities can have detrimental impacts on global ecosystem functions.

Case: Karla is a microbiologist located in Marburg. She is studying the direct effects of frequently occurring pollutants on microbial communities using microscopic imaging-based screening. Karla is especially interested in the concentration limit after which physical effects are visible (e.g. cell shape and size) within defined acquisition times. Furthermore, she wants to trace the uptake of larger waste particles and countermeasures taken by the cells including transcription profiles. Her institution is encouraged to transparently provide all data for public inspection and reuse for at least ten years.

Linked institutions and network: UMR (lead); UFZ; RWTH; JLU; IGB; NFDI4BIOIMAGE.

Challenges and NFDI4Microbiota measures:

• Determine suitable experimental setups (M2.2; M2.3).
• Convey metadata between multiple imaging experiments and omics datasets (M2.1; M3.2).
• Analyze large sets of imaging data and integrate corresponding omics datasets (M3.2; M3.8; M4.2).
• Provide long-term public access to several gigabytes of data (M2.4; M3.5; M3.6; M4.2).

5.1.7.13 Use case ‘MicroORKG’: Making knowledge from microbiological literature FAIR

Scientific relevance: As the outcome of research is primarily archived in the scientific literature, it is human-readable only and hard to query efficiently. Hence, there is a need to store statements from research articles in a systematically structured fashion.

Case: Bernd is a postdoc performing research on antibiotic resistance genes. He has compiled a large collection of scientific articles and book chapters that describe the resistance of bacterial strains from clinical isolates that show resistance against commonly used antibiotics. He now wants to share this collection and make it easily searchable.

Linked Institutions & network: ZB MED (lead); DSMZ; EMBL; JLU; TIB; NFDI4DataScience.

Challenges and NFDI4Microbiota measures:

• Research knowledge is not easily findable (M3.1, NFDI4Microbiota will host an Observatory for the Open Research Knowledge Graph *30 and help the community to include statements; M3.3; M3.6;M4.2; M4.4).
• There is a lack of awareness of linked-open-data solutions and skills for translating knowledge into machine-readable formats in the microbiological community (M1.1; M1.2; M1.3).

5.1.7.14 Use case ‘MULTI’: Integration of multi-omics data of microbial species

Scientific relevance: The integration of different types of omics data such as genomics, transcriptomics, and proteomics, as well as more specific protocols such as single-cell RNAseq for in-depth analysis andChIP- and CLIP-Seq for protein-DNA and RNA-DNA interactions, can be used to detect different cellular events. This is crucial for a comprehensive understanding of the physiology of microbes.

Case: Fiona is a microbiologist in the institute of infection biology at the University of Cologne. She studies the bacterial pathogen Pseudomonas aeruginosa and its role in cystic fibrosis in children. Her working hypothesis is that a yet poorly described transcription factor may be involved in the regulation of infection, based on transposon insertion sequencing (Tn-seq) results obtained by another group. To test this hypothesis, she has generated a knockout mutant that she used in longitudinal infection experiments in several cell lines. She now wants to integrate multiple data types: (sc)RNA-Seq, proteomics, andChIP-Seq.

Linked Institutions & network: ZB MED (lead); JLU; UFZ; BIBI; RWTH; FSU Jena; GU; HIRI; FZJülich; JMU; KIT; FLI; LMU; MLU; MDC Berlin; MPI Bremen; RUB; UK Erlangen; Uni Greifswald;TRR124; SPP2002; VAAM.

Challenges and NFDI4Microbiota measures:

• Find related datasets with similar experimental and instrumental setups (M2.1; M3.3).
• Define requirements for each data type, including quality controls, read trimming, mapping and quantification (M2.3; M3.2).
• Integration into system biological models (M3.8).
• Consistent submission of data and connected metadata to different repositories (M2.4; M3.5; M4.2).

5.1.7.15 Use case ‘PARA’: DNA modifications in gametocytes of the malaria parasitePlasmodium falciparum

Scientific relevance: DNA methylation is an important epigenetic modification that regulates gene expression. In the gametocytes of Plasmodium falciparum, which are responsible for parasite transmission from the human host to the mosquito vector, DNA (de)methylation has never been studied, althoughit is likely involved in gene regulation within this deadly parasite.

Case: Jean-Pierre Musabyimana from Aachen uses Nanopore sequencing to identify modified cytosine residues in different gametocyte stages. He now intends to use a novel DNA immunoprecipitation sequencing approach (DIP-Seq) to spot specific DNA modification sites. Whilst he feels confident with the wet lab part of the work, thanks to his training as a molecular biologist, he knows that data analysis will be a big challenge.

Linked institutions & network: RWTH (lead); ZB-MED; LMU; DGP.

Challenges and NFDI4Microbiota measures:

• Optimize DIP-Seq and implement it on experimental and clinical samples (M2.1; M2.2; M2.4).
• Analyze data from two different sequencing technologies (M2.3; M3.2; M4.1).
• Manage the data from generation to publication (M2.4; M3.5; M3.6; M3.9; M4.2).

5.1.7.16 Use case ‘PATHO’: Studying cohorts of bacterial pathogens isolated from hospitalacquired infections

Scientific relevance: According to the World Health Organization (WHO), antimicrobial resistance (AMR) is one of the biggest threats to global health, as it hampers prevention and treatment of an ever increasing range of infections caused by bacteria, parasites and fungi. Hence, it is essential to detect and monitor the occurrence and spread of pathogenic bacteria and their antimicrobial resistances.

Case: Helga is a medical microbiologist responsible for the routine surveillance of AMR in bacteriaand the management of nosocomial outbreaks within the university hospital in Frankfurt. Samples are routinely sequenced in the institute’s local sequencing facility. However, as Helga is short of bioinformatics staff and performing data analyses is a tedious task for her, NFDI4Microbiota representsan ideal solution.

Linked institutions and network: BIBI (lead); JLU; RWTH; UK Erlangen; MDC Berlin; NFDI4Health;DGHM.

Challenges and NFDI4Microbiota measures:

• Routine sample collection and processing (M2.2).
• Structured data and metadata acquisition (M2.1; M2.4; M3.4).
• Genome processing and comparative analyses, including comparisons to publicly available genomes (M3.2; M4.1; M4.2).
• Investigation of nosocomial outbreaks and comparison with other hospital sites (M3.1; M4.4).

5.1.7.17 Use case ‘PHYLOGEN’: Phylogenomics of plant-associated microbiomes

Scientific relevance: Climate change is one of the biggest challenges of our times and will severely impact agriculture and food production worldwide. One important aspect to consider is how plants and their microbiomes adapt to higher temperatures, drought, and other changing environmental conditions.

Case: Christine is an environmental microbiologist studying the composition of plant species-specific bacteria under the influence of continuously increasing surface temperature or of rising levels of atmospheric CO2 produced using free-air CO2 enrichment (FACE) technology. She needs (i) an accurate longitudinal assessment of bacterial communities based on 16S rRNA gene amplicon sequencing, and (ii) harmonized annotation, precise taxonomic classification, and phylogenomic and comparative analysis of whole genomes from selected key species. Christine is also planning to integrate her results on a higher level with data obtained from other biodiversity studies and eventually share all results with the general public to raise awareness of the impact of climate change.

Linked institutions & network: JLU (lead); DSMZ; BIBI; EMBL; UMR; GFZ; NFDI4Biodiversity;DataPLANT; NFDI4Agri; NFDI4Earth.

Challenges and NFDI4Microbiota measures:

• Routine screening of bacterial communities & quality controlled genome analysis (M2.2; M3.2; M3.5; M4.2).
• Structured data and metadata acquisition (M2.1; M2.4; M3.4).
• Robust taxonomic classification and phylogenomic analysis of fully sequenced genomes (M3.2; M3.3).
• Comparative analysis of the institute’s own genomes and public genomes across different climate conditions (M3.6; M4.1).

5.1.7.18 Use case ‘PLANT’: Root-associated microbiomes

Scientific relevance: The rhizosphere sustains one of the most diverse host-associated microbial communities, with an estimated 30,000 bacterial species. These microbes affect plant adaptation, evolution, and health and have a real impact on agricultural sustainability. It is therefore essential to study their diversity and interactions.

Case: Rosa is a PhD student in Bielefeld working on the influence of the rhizosphere microbiome on crop plant health using shotgun metagenomics, with the additional aim of integrating metatranscriptomic and metabolomic data. In the long run, she also wants to find and compare data from other studies in the context of a meta-analysis. Hence, Rosa has lots of questions regarding optimal tools, data accessibility, and required computing power.

Linked Institutions & network: DSMZ (lead); JLU; BIBI; FSU Jena; UMR; UK Essen; HMGU; MLU;MPIPZ; TUM; Uni Hamburg; NFDI4Earth; NFDI4Agri.

Challenges and NFDI4Microbiota measures:

• Choose optimal software tools and workflows (M2.2; M2.3; M4.3; M3.5).
• Estimate and optimize computing capacities for complex analyses (M3.2; M4.1).
• Access project-related data from others with rich and standardized metadata (M2.1; M3.1; M3.3; M3.4).

5.1.7.19 Use case ‘RapSARS’: Rapid access to new SARS-CoV-2 data

Scientific relevance: Rapid processing and sharing of sequencing data is crucial in research, particularly when fast responses are required as in the case of pandemic outbreaks such as COVID-19. A standardized platform for rapid data availability could be an important breakthrough for developers of new technologies, for scientists studying the biology and epidemiology of viral infection, and for health authorities.

Case: Hortense is a clinician in charge of a large COVID-19 study looking for associations between viral genomes and clinical metadata such as disease severity. NFDI4Microbiota will offer harmonized metadata management before assigning barcodes and sticking them to the samples, which are then shipped for sequencing. Sequencing data are integrated on the fly into the NFDI4Microbiota platform, which activates automated data analysis workflows and provides electronic notifications when jobs arecompleted. In parallel, Carla, a data scientist from the health authority, has submitted a request through her user interface on the platform, asking to be alerted whenever new, relevant data becomes available. Even though she does not work directly with Hortense, she can gain access to the new sequencing data via the NFDI4Microbiota platform and explore the phylogeography of the sequenced strains and theirprevalence in Germany.

Linked Institutions & network: HZI (lead); JLU; BIBI; FSU Jena; RWTH; RKI; OvGU; ZIK Septomic;Uni Köln - IMSB; Frauenhofer CIMD; NFDI4Health; GHGA; GfV.

Challenges and NFDI4Microbiota measures:

• Implement features that allow sequencing and analysis of private and public datasets in compliance with German data privacy regulations (M2.3; M2.4; M3.4; M4.1; M4.2).
• Build synergies with NFDI4Health and GHGA to provide rapid access to sequencing data and metadata as well as search functionalities, while respecting data privacy and ethical rules (M2.1; M2.4; M3.1; M4.3).
• Implement a workflow connecting sequencing centers, clinics, scientists, and health authorities (M3.1; M4.3).

5.1.7.20 Use case ‘RNA viruses’: Pathomechanisms of RNA viruses

Scientific relevance: Virus infections remain a major threat to human health. RNA viruses are of particular interest because their replication machinery introduces a high number of nucleotide substitutions, leading to enhanced adaptive behaviors. A recent example is the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which caused the current COVID-19 pandemic.

Case: Lisa is a physician working in pulmonary research. She is currently investigating the pathomechanisms of virus-induced lung injury, especially in patients infected with SARS-CoV-2. Her methods include analyzing the transcriptome of both host cells and the virus in time-series in vitro experiments by combining bulk and single-cell RNA sequencing. This results in large, complex datasets, for which multiple bioinformatic pipelines are required, which are out of her area of expertise.

Linked institutions and network: FZU Jena (lead); JLU; HHU; SFB0221; Frauenhofer CIMD;NFDI4Health; GfV.

Challenges and NFDI4Microbiota measures:

• Standardize experimental design for bulk and single-cell RNA-seq (M2.2).
• Routinely collect clinical samples of virus-infected host cells (M2.4).
• Structured data and metadata acquisition (M2.1; M3.4; M4.2).
• Standardize data processing & comparative analyses (M3.2; M3.8).
• Combine analysis of data from several hospital sites and epidemiology (M3.1; M4.4).

5.1.7.21 Use case ‘SMALLPRO’: Prokaryotic small PROTEINS in gut metagenomes

Scientific relevance: Less than half of all proteins produced by gut microbes can be annotated, not to mention those with false annotations. Specialized small proteins encoded by Biosynthetic Gene Clusters (BGCs) represent an important, understudied group of molecules within the gut microbiome, since their biosynthesis is usually a multi-step process and they may have relevant functions such as anti-microbial activities. It is therefore important to explore gut metagenomes to provide detailed characterization of such molecules.

Case: Jaymi, a PhD student in Aachen, has established multiple collections of bacterial strains from the intestines of humans and other animals. In collaboration with TU Dresden and SPP2002, she identified a potentially new family of BGC-encoded small proteins in the genome of some of her strains. NFDI4Microbiota will be able to help her take this project further, e.g. by screening the genomes of all other strains for additional, similar discoveries and by performing targeted investigations of the small proteins she has already identified.

Linked institutions & network: RWTH (lead); EMBL; HKI; SPP2002; SFB1371; SFB1382;NFDI4Biodiversity; VAAM, DGHM.

Challenges and NFDI4Microbiota measures:

• Genome sequence processing and curation (M2.1; M2.4; M3.2; M3.5; M3.6; M4.1; M4.2).
• Quantitative measurement of identified small proteins by proteomics (M2.2; M2.3; M3.9).
• Impact of the target small molecules on the gut microbiome in vivo by multi-omics (M2.2; M3.8).

5.1.7.22 Use case ‘Soft-DEV’: Scientific software developer

Scientific relevance: Contemporary bioinformatics software development often requires the integration of various public and private data repositories. Thus, the storage platform of NFDI4Microbiota will support scientific software developers by providing access to data and metadata via highly standardized frameworks.

Case: Lev is a software developer who designed a new algorithm which requires the analysis of metadata. He can access the metadata stored along with biological data using the standardized REST API provided by NFDI4Microbiota and further process this metadata using the well-documented and easily accessible JSON format. Further, large-scale analyses are supported by the storage platform’s high-performance backend. Finally, Lev is able to directly integrate the REST interface into his new applications, allowing users to easily access the data. An authentication layer can be used to restrict access and upload to the platform using ELIXIR AAI.

Linked institutions & network: JLU (lead); BIBI; EMBL; FSU Jena; GFZ; FLI; RSE4NFDI.

Challenges and NFDI4Microbiota measures:

• Train people to use the platform (M1.1).
• Build a scalable storage platform (M3.1; M4.2; M4.3; M3.5).

5.1.7.23 Use case ‘SPACE’: Microbial isolates from outer space

Scientific relevance: Public and private space organizations are eager to resume crewed space travel targeting the moon and Mars. Investigating microbes within the closed habitat of spacecraft during longer space journeys is important, as exposure to harsh space conditions (high levels of UV and ionizing radiation, space vacuum, low gravitational force) may drive harmful mutations.

Case: Juri is a microbiologist at the DLR studying bacterial isolates obtained from space missions. He is responsible for monitoring bacterial and viral contamination based on swipe samples from the ISS and hardware assembly facilities of the European Space Agency (ESA). These samples, and the single microbes they contain, are routinely sequenced and need to be analyzed in a standardized manner, forwhich Juri urgently needs support.

Linked institutions & network: DLR Cologne (lead); JLU; FSU Jena; UMR; GFZ; PUNCH4NFDI.

Challenges and NFDI4Microbiota measures:

• Routine collection of microbial samples and preparation for sequencing (M2.2; M2.3).
• Structured data and metadata acquisition (M2.1; M2.4; M3.3; M3.5).
• Genome processing and comparative analyses with third-party data (M3.2; M3.4; M3.6; M4.1; M4.2).

5.1.7.24 Use case ‘Strain-ID’: Collecting and matching microbial strain identifiers

Scientific relevance: Strain identifiers represent a universal language for scientists, clinicians and industry, as well as for legal compliance and regulatory control purposes. They enable users to generate and communicate findings and to compare and verify data, ensuring traceability and reproducibility. Whilst unique identifier systems exist (INSDC accessions, E. C. enzyme numbers), microbes are referenced only via numbers issued by culture collections and informal historical identifiers.

Case: André is a postdoc in mycology who inherited a strain collection of biotechnological interest from a retired colleague. Unfortunately, due to inconsistencies in strain identification in the literature, it is difficult for him to appreciate the value of his strains. He remembers that the database straininfo.net may be useful, but discovers it has been put permanently out of commission. With >700 culture collections worldwide using different collection numbers, the task of tracing and obtaining reference strains will be a laborious one.

Linked institutions & network: DSMZ (lead); RWTH; Uni Göttingen; TRR124; NFDI4Biodiversity;DGfM.

Challenges and NFDI4Microbiota measures:

• Build a database that collects and matches strain identifiers from different sources and provides stable referencing as well as linking with other identifiers, e.g. INSDC sequence accession numbers (M2.1; M2.4; M3.1; M3.3; M4.3; M3.5).

5.1.7.25 Use case ‘SYNSYS’: Synthetic systems

Scientific relevance: Polystyrene is an extremely recalcitrant pollutant accumulating as litter in the environment. Developing sustainable and scalable methods to degrade this pollutant is of utmost global relevance.

Case: Philipp, a synthetic biologist from Marburg, works with Denise, a microbial ecologist from Leipzig, on microbe-based degradation of polystyrene litter. They aim to implement a synthetic pathway for digestion that is scalable to large-scale industrial use. Their collaborative project will include both laboratory and computational work and will be carried out at two separate locations with different backgrounds but entangled objectives. Proper harmonization and sharing of data is therefore essential.

Linked institutions and network: UMR (lead); UFZ; ZB-MED; JLU; HMGU; FZ Jülich; IGB; MPIMagdeburg; TU Darmstadt; NFDI4Cat; NFDI4Chem; NFDI4Ing; VAAM.

Challenges and NFDI4Microbiota measures:

• Screen public omics and text data for relevant enzymes for pathway construction (M2.1; M3.1; M3.4; M3.7).
• Determine adequate workflows for genetic construction, testing and optimization (M2.2; M2.3).
• Evaluate and integrate omics and imaging data including datasets from third parties (M3.2; M3.8).
• Manage, organize, and share large amounts of data (M2.4; M3.5; M3.6; M4.2).

Fig. 8

Contributions: Of all the task areas, TA1 brings together the highest number of contributions by all the co-applicants, just ahead of TA3 on services. This clearly demonstrates how – aside from the technical aspects – NFDI4Microbiota places a great deal of importance on activities linking this consortium to the community and beyond via training, support, outreach, and collaboration. Contributions by the lead and co-lead (HZI and ZB-MED) are highest, followed by RWTH (all three institutions with more than 100PM) for coordination of the use cases, which are key instruments for bringing all stakeholders together around collaborative scientific projects. The contribution of all other co-applicants averages around 60PM/institution. Together, the co-applicants provide extremely solid foundations for this TA through their in-kind contributions – in the form of a multitude of existing training and support activities – and through their networking with multiple participants, other NFDI consortia, and international partners. Forthcoming contributions within NFDI4Microbiota will include coordinating and harmonizing existing measures to boost them further, as well as developing new measures to stay at the forefront of a rapidly moving field of research.

Dependencies: The success of TA1 is intrinsically linked to the entire landscape of microbiota research in Germany, since NFDI4Microbiota will act as an important interface between the co-applicants, who bring together a rich portfolio of expertise, and a community that represents a multitude of needs. Building bridges between the consortium and this community is the biggest dependency of TA1, but also its biggest strength, since the consortium will benefit from community feedback in return. This task area also relies on all the other TAs and measures implemented by NFDI4Microbiota, since training, support, and use cases are catalysts of community interaction based on all the various tools, measures, and services developed by the consortium. This broad scope of dependencies is, in fact, a strength, as progress within TA1 will not be hampered by unexpected events that may slow down the progress of one specific arm of the technical working program.

Risks of implementation and mitigation approaches: (1) There is a risk that the target communities will not be strongly engaged by NFDI4Microbiota. However, we think this risk is low, because we developed the concepts for this infrastructure in close collaboration with the community and consistently received very positive feedback. Continuing to keep the community actively involved is one of the consortium’s most important goals, as illustrated by its dedicated training and outreach plans. Should this not suffice, we will use our flexible funds (M5.3) to invest further in community outreach and engagement and consider their feedback even more extensively in all consortium activities, as well as further leveraging our own contacts within these communities. (2) Another risk is that the demand for training could exceed available capacity since, in our experience, the demand for training in computational methodologies is usually very high, with double or triple overbooking. Should this happen, we will allocate more flexible funds to training, prioritize training courses and participant selection inconsultation with the SAB, and increase our focus on creating reusable, scalable online training resources and making use of existing high-quality national and international training resources. (3) Further, it may prove difficult to identify adequate strategies for guaranteeing sustainability and long-term maintenance, as this is a major challenge for all infrastructures. However, the consortium members are all very experienced and well-connected, so we believe suitable long-term solutions will be found. We also firmly believe that the extended training we are planning will help make this infrastructure a success and guarantee its long-term usage.

Contributors: UFZ (lead, 69 PM), FSU Jena (co-lead, 19 PM), DSMZ (6 PM), EMBL (4 PM), UMR (4 PM), RWTH (8 PM), JLU (6 PM)

Goals: To maximize the quality of data entering the NFDI4Microbiota system by enforcing compliance with existing standards, as well as to identify and promote additional tailored data standards and metadata requirements within the NFDI4Microbiota systems.

Pipelines that use data from omics or state-of-the-art microbiota analysis rely on the quality of the data and the completeness of the metadata. The diversity of data types, how they are produced and the different metrics used to assert their quality are not always clear, particularly for non-experienced users. As data quality is of the utmost importance for any subsequent analysis, this task area will enforce compliance with existing standards and, where necessary, define further standards for the various techniques relevant to NFDI4Microbiota in collaboration with international initiatives (M1.5 - Connection to international partners). We will use the commenting/reviewing services (M3.7 - Reviewing/commenting system for data) to improve the interface used for data quality for different domains of data. We will also request feedback from scientists at different career levels and adjust which types of data are relevant. The standards employed by the NFDI4Microbiota consortium will ensure that only high-quality raw data entered together with sufficient and standardized metadata is analyzed by the NFDI4Microbiota resources. Further, standardized metadata will facilitate future integration and comparison of the data with novel state-of-the-art analyses or new data. Measure M2.1 will define open, easy-to-use interfaces to check if raw data and metadata meet the quality standards desired by the NFDI4Microbiota Networks and will provide users with access to these interfaces. To define the standards for raw data quality, we will use the capacity of different co-applicants and participants specialized in the different types of data. Metadata standards will be categorized as technical (i.e. dependent on the technology used for the data), biological (i.e. dependent on the microbiota analyzed) or environmental (i.e. defining the environmental variables relevant for those samples). The technical metadata will be defined by the various data-type specialists within the NFDI4Microbiota Network while the biological and environmental metadata will be defined by specialists from various fields of the consortium (e.g. human microbiomes, terrestrial microbiome and marine). We will enforce compliance with standards defined by key initiatives such as: the Genomic Standards Consortium (Field et al. 2008), which defines the minimum necessary information on an omic sample (Yilmaz et al. 2011), which contextualizes biological and biomedical entities; and the OBO Foundry (ref), which coordinates the evolution of ontologies to support biomedical data integration. The NFDI4Microbiota team has already worked on metadata standardization for metagenomes from terrestrial environments (Corrêa et al. 2020). To avoid the bottleneck of users having to search for all the information required for their specific data and metadata once the standards for data and metadata are defined, we will generate user-friendly applications to enable users to check the quality of their data and metadata. Therefore, Measure M2.1 will link data generation and downstream analysis by creating materials that train users how to measure the quality of the various data types supported by the NFDI4Microbiota consortium. The development and definition of metadata standards will be carried out in close cooperation with M4.2 - Data storage platform to ensure efficient metadata handling in the storage backend. Table 16

Contributors: EMBL (lead, 41 PM), DSMZ (co-lead, 3 PM + sig. in-kind), RWTH (12 PM), FSU Jena (12 PM)

Goals: To increase the uniformity and interoperability of data collected within the NFDI4Microbiota network by identifying and promoting best practices for performing microbial research.

A major source of biological variation in experimental results comes from differences in how experiments are performed and how biological materials are collected and processed. While statistical methods exist to counteract these barriers to data integration, and will be made available to users through M3.2 - Analytical services, these approaches are not always sufficient to overcome data incompatibilities. The centralization of information in the NFDI4Microbiota network provides the opportunity to streamline and unify experimental procedures from the start, which would improve future meta-analysis outcomes (M3.8 - Systems biology, modeling, and multi-omics integration), prevent duplication of effort in developing SOPs, and spread expertise between fields (M1.4 - Connection to other NFDI consortia). This measure will provide users with an open, version-tracked resource of curated standard operating procedures (SOPs) for experiment design, sample collection, storage, and processing in culture-based and microbiome studies. These SOPs will provide guidance at both a general level (e.g. sampling of any microbial community) and for specific types of starting material (e.g. sampling microbial communities from human faeces). SOPs will be collected from NFDI4Microbiota participants, existing initiatives (e.g. IMHS (www.microbiome-standards.org/), the Earth Microbiome Project (https://earthmicrobiome.org/), the European Microbial DNA Bank Network (www.microdnabank.eu/), and peer-reviewed publications. The NFDI4Microbiota SOP resource will improve on existing resources by curating and integrating SOPs from distinct experimental approaches to encourage the production of integratable data types. Initial SOPs selected will be for well-established experimental approaches. As best practices evolve and new experimental techniques mature, SOPs will be added and updated, guided by the NFDI4Microbiota Ambassador Council. SOPs will be provided to the community in the centralized NFDI4Microbiota web system M3.1 - Central web portal. This will support commenting, which will facilitate community dialogue around the SOPs, revisions, which will ensure SOPs can evolve with new best practices, and version-tracking, which will support referencing of exact SOPs, thereby facilitating reproducible research. Table 17

Contributors: HZI (lead, 60 PM), BIBI (co-lead, 27 PM), JLU (4 PM), RWTH (8 PM), EMBL (3 PM),FSU Jena (12 PM)

Goals: To develop guidelines for standardized and FLOSS-compliant workflows, and to establish standard workflows for data quality control and benchmarking services. To create maximum value for the community, NFDI4Microbiota will offer standardized workflows for users to analyze their own datasets as well as running these on curated datasets created and provided by the consortium. Work in this task area will focus on developing best practices for standardized workflow design and development, as well as for microbiome data processing (e.g. using Common Workflow Language (CWL), Snakemake, Nextflow). The consortium will ensure that (i) workflows are developed according to FLOSS principles (Free/Libre/Open Source Software), (ii) code is accessible, (iii) software is versioned with stable releases, (iv) software is containerized (e.g. using Docker) to ensure reproducibility of results across different computing environments, which otherwise might differ, and (v) the application programming interfaces (APIs) are made public. In addition, data provenance will be tracked within the workflows. We will establish joint working groups with national and international partners (e.g. EMBLEBI MGnify, EOSC *24, the ELIXIR Interoperability Platform and de.NBI) to develop international standards in workflow design, e.g. emphasizing a modular approach to enable exchange of workflow parts. We will also implement strategies for workflow maintenance and performance optimization (e.g. runtime and memory consumption) and ensure sustainability and high quality standards for implemented workflows by emphasizing documentation and automatic testing. The EDAM ontology (Ison et al. 2013) will be used to define types of data and data identifiers, data formats, operations and topics for the established workflows. EDAM provides a set of concepts with preferred terms and synonyms, definitions, and additional information.

To complement quality control performed by data stewards, we will establish standardized quality control workflows for data including sequencing data, data processing results and metadata, such that all data held by the NFDI4Microbiota platform complies with the data and metadata standards established in M2.1 - Data & metadata standards. The created workflows and results will be strictly versioned according to the principles of semantic versioning, indicating minor (e.g. version 1.0.1 to 1.0.2) and major software changes (e.g. version 1.1.0 to 1.2.0). Data generated with different releases of the same software will be labeled accordingly, reinforcing comprehensibility and reproducibility of findings based on data processed with the NFDI4Microbiota platform. We will also develop a process for integrating user-defined, standard-compliant workflows into our system (together with M4.4 - Service monitoring & reporting). These will undergo a quality review process before being offered as an option to the community.

The computational services offered, i.e. the functionalities available within workflows, will be based on user requirements (e.g. as determined in M1.3 - Community outreach and public relations) and bestpractice recommendations determined in systematic benchmarking efforts, such as ELIXIRs OpenEBench platform and CAMI, the Initiative for the Critical Assessment of Metagenome Interpretation (Sczyrba et al. 2017), organized by A.C. McHardy and A. Sczyrba. The infrastructure will thus offer users the most suitable approach for each task, so that they can perform analyses of their data without needing to be experts in the plethora of computational methods used in that area. Implementation of specific functionalities will be prioritized by expected user demand and community relevance.

Where indicated, we will benchmark methods for individual use cases (e.g. workflows for depositing and analyzing clinical microbiome datasets in interaction with GHGA, M1.7 - Use cases). To this end, we will provide benchmarking options and a representative collection of metagenome benchmark datasets, enabling users to assess software for computational microbiome research. This service will build on community-derived benchmarking concepts developed e.g. in CAMI and ELIXIR, such as metrics and visualizations embedded in benchmarking software, Docker-based standardized virtualizations of tool categories (e.g. metagenome assemblers) in bioboxes (Belmann et al. 2015) and datasets (Meyer et al. 2021, Sczyrba et al. 2017). This will facilitate the comparison of developed methods with existing approaches and the development of performanceoptimized analytical workflows.

Communication of developed standards and best practices will be connected to (M1.3 - Community outreach and public relations; M1.1 - Training and education). Table 18

Contributors: UMR (lead, 48 PM), JLU (co-lead, 2 PM), RWTH (12 PM), DSMZ (6 PM)

Goals: To develop and implement policies for data management, sharing and reuse that ensure quality and openness of data but also legal certainty. Our policies will respect the interests of users, provide guidance on making responsible use of available resources, and safeguard compliance with the FAIR principles, thus stimulating the reuse of data and fostering new research.

NFDI4Microbiota offers a broad range of guidelines, integrated tools, and workflows for the preparation, processing, and analysis of research data (M2.1 - Data & metadata standards, M2.2 - Experimental procedure standards, M2.3 - Workflow standards, M3.2 - Analytical services, M3.8 - Systems biology, modeling, and multi-omics integration). This data needs to be evaluated, stored, organized, secured, backed up, shared, and, where necessary, safely removed. For this purpose, data management plans will be created and managed using the Research Data Management Organizer (RDMO) (Neuroth and Engelhardt 2018) tool. These plans will also cover all intermediate files, result files, and visualizations that are generated in the process of an analysis. The management plans will be translated into applicable policies that are then implemented in the workflows and, wherever possible, applied automatically within implemented procedures and workflows. Compliance with policies will be supervised and monitored regularly by internal implementation reviews and functional real-life tests where actual data will be used that might violate current policies. The initial stage will be a concept of management plans and policies that follow internationally established recommendations and guidelines on data sharing e.g. by the Research Data Alliance (RDA) (RDA COVID-19 Working Group 2020). Based on this concept, we will acquire feedback from potential users, participants, and members to compile a set of definitions that is generally agreed upon within the consortium. This process will be iterated several times during the progression of NFDI4Microbiota in order to cover the latest needs and developments, to harmonize with other NFDI consortia such as NFDI4Biodiversity, DataPLANT, NFDI4Health, NFDI4Earth (M1.4 - Connection to other NFDI consortia) and international partners (M1.5 - Connection to international partners), and to maintain interoperability with major data providers such as the centers of the Next Generation Sequencing Competence Network (NGS-CN), EBI, NCBI and UniProt, whose usage we explicitly foster.

NFDI4Microbiota will develop a framework of operational principles and guidelines that ensures open access, reproducibility, consistency, transparency, and interoperability throughout the whole data workflow. Compliance with FAIR principles will be supported by developing all platform components as free and open source software (FOSS) within our NFDI4Microbiota consortium. This includes the application of best-practice guidelines for software engineering and community interaction, as well as the use of OSI-compliant licenses. To foster collaboration with other NFDI consortia, we will ensure that all development processes are transparent to outside users and software developers. Generally applicable guidelines ensuring open access, reproducibility, consistency, transparency, and interoperability will be embedded in the training (M1.1 - Training and education).

Legal issues are an overarching topic that needs to be addressed as a cross-topic by the whole NFDI (Leipzig-Berlin declaration (Bierwirth et al. 2020)). Nevertheless, microbial research poses specific challenges that need to be considered from the start (e.g. how to handle remnant/trace amounts of host genomic information, risk classification). NFDI4Microbiota will therefore be at the forefront of efforts to educate and support microbial researchers. This includes the incorporation of legal issues and how these should be handled in training (M1.1 - Training and education).

The legal conformity of data and metadata will be ascertained upon submission. NFDI4Microbiota will ensure that all work performed within the consortium complies with the legislation in force that governs the use of personal health information data, including the World Medical Association (WMA) Declaration of Helsinki and the WMA Declaration of Taipei. In addition, the General Data Protection Regulation (GDPR, EU-R 670/2016) will be considered in light of additional national or state-specific regulations. Conditions for processing and sharing by NFDI4Microbiota will be sufficiently anonymized and, if applicable, additional data filtering will be applied (e.g. removal of contaminant human sequence fragments from shotgun metagenomes). The TA leaders will coordinate the structures and processes of NFDI4Microbiota with regard to ethical and privacy issues in close cooperation with the Technologie und Methodenplattform für die vernetzte medizinische Forschung e.V. (TMF) and other NFDI consortia such as GHGA and NFDI4Health (M1.4 - Connection to other NFDI consortia). Legal conformity will be re-assessed regularly in case new methods arise that could re-enable the traceability of individuals.

The Nagoya Protocol *34 is a legally-binding international instrument that came into force on October 12, 2014 and established procedures for access and benefit-sharing. For scientists working internationally – even within different EU member states – this means that there are new legal requirements in place when accessing (sampling) or using biological material. Although Germany is a so-called free access country, which means there are no Nagoya obligations on German material, German scientists are required to follow the laws of other countries and once Nagoya-relevant documents have been obtained, these need to stay connected to the organisms/material.

The implemented policies will be strict enough to ensure legal conformity, high-quality data and adherence to the FAIR principles (Wilkinson et al. 2016), but this should not prevent researchers from using the platform. User acceptance is an equally crucial aspect that must be addressed. A typical outcome will be that data is not publicly accessible to anyone except explicitly mentioned collaborators before the research question is answered or the respective publication is finished. Even if it seemingly contradicts the idea of open science, temporary data locks and limited access to sensitive data will play a necessary part in making data accessible in the first place. Similarly, our data management policies will ensure the data produced by our users has a long life cycle, making it easily findable and widely accessible and helping to foster new research. Table 19

Fig. 10

Contributions: Reflecting the importance of TA2 and the expertise that all co-applicants bring, 9/10 co-applicants will contribute to this TA. Further, participants will be encouraged to contribute to and comment on some standards and policies. Major leadership roles will be taken on by UFZ for data and metadata (with support from BIBI), EMBL for experimental procedures and overall coordination (with support from DSMZ), HZI for workflows, and UMR for data management and reuse. Several institutes will support most measures, facilitating their connectivity (DSMZ, EMBL, FSU, JLU, RWTH).

Dependencies: The creation and curation of standards and of an online resource to disseminate them to the microbiological community will be supported by the in-depth microbiological, ethical and research expertise of the co-applicants and of the participants. Establishing standards is a moving target, as standards may evolve over time as the field evolves. This dependency on ongoing developments is mitigated by the strong ties that NFDI4Microbiota has to the multiple stakeholders in the field, which will ensure it is made aware of any changes that may need to be made to standards. The development and use of the online resource will depend on the web interface (M3.1), and, importantly, on training, support, and raising awareness among users (M1.1, M1.2, M1.3).

Risks of implementation and mitigation approaches: The major risk to the success of this Task Area concerns the uptake of and adherence to these standards and policies, which will depend on the engagement of users. The risk of user disuse or misuse will be minimized through provision of comprehensive training, support and outreach across Germany (M1.1, M1.2, M1.3). This will occur both through the organized, official activities of NFDI4Microbiota, as well as through community relationships, which are significant given the wide coverage of institutes and fields among NFDI4Microbiota’s coapplicants and participants, and the emphasis placed on community building (M1.3). A secondary risk involves disagreements arising in the selection or promotion of some standards over others. This will be mitigated by taking an inclusive approach to the presentation of standards and supporting a user forum in which benefits and limitations can be openly discussed. The use of respectful and inclusive communication will be enforced.

Contributors: ZB MED (lead, 60 PM), EMBL (co-lead, 18 PM), DSMZ (6 PM), JLU (4 PM), UFZ (12 PM)

Goals: To provide numerous and wide-ranging services to the microbiology community in Germany and beyond. These will be offered through a one-stop shop, a central web-based portal known as the NFDI4Microbiota Hub.

Integration of services: The portal will act as an information resource while also giving direct access to the services NFDI4Microbiota provides to the community. For this purpose, it will integrate the various internal analytical services (M3.2 - Analytical services), collect analysis results, link to external resources, provide access to data, and enable users to search for datasets and query across multiple databases and services (M3.3 - Databases & terminology services). It will also provide connections to tools and resources not (yet) included in the NFDI4Microbiota platform by linking to existing service listings, such as bio.tools *35.

Furthermore, it will allow users to sign up for training events (M1.1 - Training and education), provide access to help desks (M1.2 - Support), and serve as a marketplace to bring together researchers with complementary expertise and list job vacancies.

Subportals: Having a single, central portal will create an anchor point for the diverse communities brought together by NFDI4Microbiota’s activities. However, to facilitate usage of domain-specific tools and avoid confronting users with overwhelming options, we will also create tailored portals adapted to the needs of NFDI4Microbiota sub-communities, such as COVID-19 researchers, cell culture specialists, and bioengineers. These tailored portals will maintain clear ties to the central portal through their similar look and feel and consistent structure. Once users are comfortable using the website within their domain of expertise, they can easily move on to other areas of the website.

Search: The ability to search for datasets efficiently is crucial. The NFDI4Microbiota Hub will include a powerful discovery service that will also enable users to search based on biological and/or experimental setups. Literature search functions will be integrated through ZB MED’s discovery service LIVIVO *36 with its underlying ZB MED Knowledge Environment and state-of-the-art semantic search technologies.

API: Besides the web interface, the portal will also provide a rich API for querying metadata and performing automated bulk uploads and downloads. It will feature a user-friendly front-end for data submission (building on the data deposition back-end developed in M3.5 - Data deposition & repositories) and allow users to curate metadata directly within the web interface. Table 20

Contributors: HZI (lead, 90 PM), BIBI (co-lead, 27 PM), JLU (18 PM), EMBL (16 PM), UMR (5 PM + sig. in-kind), FSU Jena (21 PM), RWTH (12 PM)

Goals: To develop and maintain best-practice analytical workflows and services for processing microbiological research data.

Fundamental to the aims of the NFDI4Microbiota consortium is the development and provision of the computational infrastructure and analytical workflows required to store, access, process, and interpret various microbiome-related data types. This data may be generated by a range of use cases (M1.7 - Use cases) and may include omics data of microbial communities, bacterial and archaeal isolates, as well as fungi, parasites and viruses (see M3.3 - Databases & terminology services). To develop workflows, NFDI4Microbiota will establish synergies through international cooperation, making use of reproducible, open source solutions wherever possible (e.g. *24), and interacting, for instance, with working groups of the ELIXIR Interoperability Platform and teams working on EMBL-EBI’s MGNify database, JGI IMG Integrated Microbial Genomes & Microbiomes, MG-RAST, and the Qiita platform (Gonzalez et al. 2018) (R. Knight, UCSD). We will also carry out methodological benchmarking in community efforts such as CAMI (M2.3 - Workflow standards). Workflows, query interfaces and visualization implementations will be prioritized based on estimated user numbers and range of hosted data types. Required methodologies include software for quality control, data processing, statistical analyses, and visualizations of different data types and results.

Innovations and missing software components will be developed and integrated by the NFDI4Microbiota partners as needed, following good software engineering practice and according to the FLOSS principles (Free/Libre/Open Source Software). Further high-quality workflows developed by users will be allowed to run within the framework, providing they meet the standards established in M2.3 - Workflow standards. All developed software will be made public under Open Source Initiative (OSI)-compliant licenses. NFDI4Microbiota will cooperate with the RSE4NFDI for this purpose.

NFDI4Microbiota will also work with international partners to establish mechanisms for sharing, integrating and combining user-relevant workflow outputs such as metagenomic assemblies and MAGs, and depositing them in ENA. To ensure a high-quality user experience and excellent support, we will work together with M1.2 - Support on establishing a help desk and a user forum. The user forum will enable users to support and interact with each other and serve as a community-based knowledge resource. Additionally, we will work with M1.1 - Training and education to offer dedicated training courses on the workflows we develop, since we believe that training is key to guaranteeing active use of the NFDI4Microbiota infrastructure and ensuring its success. Tables 21, 22

Contributors: DSMZ (lead, 120 PM), FSU Jena (co-lead, 60 PM), RWTH (10 PM), UMR (2 PM), EMBL (4 PM), ZB MED (14 PM), UFZ (sig. in-kind)

Goals: To support FAIRification in database development, safeguard knowledge in databases, and build up central databases for collecting and matching strain identifiers as well as for non-pathogenic virus data.

Support FAIRification in database development and safeguard knowledge in databases: Efficient access to comprehensive and standardized data is essential for large-scale analysis. Modern highthroughput methodologies are generating data at ever-increasing rates. Tremendous amounts of data are being collected, yet there is a risk that much of it will never be reused. To make this data accessible, databases have been developed to collect and standardize scientific data. Except for a few centrally managed databases, most of these lack long-term funding. The majority of databases are developed as part of a project or a PhD thesis and are orphaned when the scientist responsible leaves. Although many of these databases are valuable, they often become inactive and are ultimately lost.

NFDI4Microbiota will establish a help desk to support researchers during all phases of a database’s life cycle (M1.2 - Support). This will include advising researchers during development, supporting them during database operation, and ascertaining when a database is no longer being actively developed. During development, many aspects of standardization and interoperability can be optimized, greatly enhancing the value for the scientific community.

Content of orphaned databases needs to be preserved, e.g. by relocating databases to a centrally managed system. In the simplest scenario, data is transferred to a long-term archive as a data dump, described with rich metadata, and thus made findable for researchers. A more sophisticated approach is to transform the data and transfer it to a public database (e.g. Wikidata) where researchers can directly access and reuse the data. The ultimate goal is to successively transfer data into a machineinterpretable format such as the Open Research Knowledge Graph (www.orkg.org) to improve access and enable semantic queries.

To ensure and enhance data findability, NFDI4Microbiota will assist researchers by providing advice (M1.2 - Support), offering training (M1.1 - Training and education), and carrying out active development to make sure that all data sources are available through the central web portal (M3.1 - Central web portal).

Central database for collecting and matching identifiers for prokaryotic and eukaryotic strains: Terminology services are of great importance for the NFDI as a whole, since unified, consistent identification and understanding of entities provides the basis for FAIR data. In the case of microbial data, terminologies are partially covered by the NFDI4Biodiversity terminology services, and NFDI4Microbiota will support an NFDI-wide terminology service. Still, there are gaps in the terminology of microbial research that need to be addressed.

One very significant gap is the lack of unique and stable identifiers for microbial strains, which are a prerequisite for the comparability and reproducibility of microbial data. The lack thereof therefore represents a major obstacle to FAIR microbial research (Use case ‘Strain-ID’: Collecting and matching microbial strain identifiers).

NFDI4Microbiota will build a database to collect and match strain identifiers from different sources. This will offer (1) a one-stop-shop for strain identifiers searchable via the central web portal (M3.1 - Central web portal) and (2) a Web service (API) for the automated retrieval of strain identifier information. (3) Constant efforts will be made to curate and match strain identifiers as well as to explore and integrate new sources including strain identifiers representing non-culturable strains. Up-to-date information will be published every six months. (4) Moreover, this service will be closely coordinated with relevant activities in other NFDIs, especially NFDI4Biodiversity.

Central virus database: Viruses are of utmost importance for NFDI4Microbiota (M1.7 - Use cases, e.g. Use case ‘GUT’: Gut microbiomes and Use case ‘METASOIL’: Mechanisms of antimicrobial resistance and degradation of pollutants in soils). The storage of virus genome data is essential and is currently a computationally unsolved problem. In future, we will need novel, qualitatively different computational methods, possibly pan-genomes to reflect the virus quasispecies as a cloud of viral haplotypes. Currently, besides the general NCBI database, a few virus-specific databases exist, such as the Virus Pathogen Resource (ViPR) *40 with a restricted focus on human pathogens, GISAID EpiFlu for Influenza A with restricted access to data, HIV DB, HCV DB, and Viralzone as the comprehensive encyclopedia for viruses without sequence storage. The surprising lack of a comprehensive viral database is the rationale behind this measure. NFDI4Microbiota is invited to use the data structure, established tools, and resources of ViPR to build a complementary virus database for non-human viruses and non-pathogenic viruses (M1.5 - Connection to international partners). The two databases will share the same underlying structure and tools, and will update each other regularly without holding content in several places and while adhering to data and metadata standards (M2.1 - Data & metadata standards). EMBL-EBI is planned as the storage location for the central virus database. This database will benefit substantially from a (non-trivial) community acceptance with an established high service grade, including several tools and web portals (M3.1 - Central web portal). Additionally, usage training and education will be offered (M1.1 - Training and education). Table 23

Contributors: UMR (lead, 44 PM), DSMZ (co-lead, 3 PM), UFZ (sig. in-kind)

Goals: To foster qualified and interpretable research by ensuring data quality and well-documented data provenance.

Data quality is a major prerequisite for qualified, interpretable, and valid research results *37. Another often underrated aspect is data provenance, which allows the tracing of data sources and processing steps. Well-documented provenance helps to confirm the authenticity of data and allows researchers to assess the comparability of studies. These aspects become even more pressing in the light of open science, which promotes reproducibility, reuse, and sharing of data. Ensuring full transparency by monitoring quality and keeping track of provenance is a resource-heavy and time-consuming task that, moreover, requires knowledge of best practices. This can become a significant obstacle to research and research groups. NFDI4Microbiota will therefore provide intuitive services to address data quality and provenance throughout the processing and analysis workflows.

Based on the type of data (M2.1 - Data & metadata standards) and the chosen workflow (M2.3 - Workflow standards), data quality will be evaluated automatically when the data is initially uploaded and after each processing step. A brief quality summary will be stored in the metadata to ensure full transparency. The detailed report can be reproduced at any time as per our consistency and reproducibility policy (M2.4 - Policies and legal issues for data management and reuse). Reporting will focus, in particular, on significant problems. Typically, users are left to cope with quality warnings on their own. In contrast, our platform will give them access to expert experience and enable them to take advantage of defined experimental procedures (M2.2 - Experimental procedure standards) and workflows (M2.3 - Workflow standards & M3.2 - Analytical services). The interpretation of quality warnings will be supported by the clarification of potential implications for the interpretation of any results and best practices. Where possible, expert advice will be given on how to proceed or step back and improve the data. This “feedback assistance” offer will be extended regularly based on frequent use cases and requests. Furthermore, the results of feedback and advice will be integrated in the respective training (M1.1 - Training and education). Most quality checks will be mandatory to ensure compliance with our quality standards.

Provenance data will be kept as metadata in the form of a knowledge graph. We will set up a framework that transfers and extends metadata along the process of processing and analyzing datasets. Processing steps, filters, tool versions and parameters, quality checks, and so on will be documented automatically without requiring any user interaction. If a user decides to modify a processing step due to the outcome of quality control, this will be noted and reasoned in the metadata as well. When using large workflow pipelines, provenance data can easily become complex and hard to follow. Thus, our service also includes a means of inspecting provenance data in the form of graph-based visualizations. Moreover, this data will be used to automatically generate a “materials and methods” text for the given dataset which will describe the processing and analysis steps in a human-readable form and greatly facilitate subsequent manuscript preparation. This text will include references to tools, databases and data sources, thereby ensuring seamless documentation of the work and supporting FAIR citation practices. Table 24

Contributors: JLU (lead, 30 PM), UFZ (co-lead, 20 PM), ZB MED (12 PM)

Goals: To develop and provide standardized interfaces and an extensible platform for submitting data to, and searching for data in, public repositories.

Data repositories such as the European Nucleotide Archive (ENA) are a valuable resource for facilitating standardized deposition and reuse of generated data. In addition, these repositories can be used as an off-site backup platform and to provide widely recognized digital object identifiers (DOIs). Submitting data to these repositories, however, is a time-consuming and non-trivial task. In many cases, this hinders researchers from depositing their data into such a repository even after their study has finished, especially if their project did not require such a submission.

To facilitate the process of data submission, this measure will provide a platform to easily submit data to a set of well-recognized repositories using a REST API. This API can be used to directly submit data from our storage system, which is developed in M4.2 - Data storage platform, to an appropriate repository. The system will handle all the repository-specific aspects of the submission and check if the data and metadata is correctly formatted and the metadata contains all necessary information for a submission. The API can be accessed either directly using ELIXIR AAI, or via the central web portal developed in M3.1 - Central web portal. Depending on the user’s preference, the data can also be deposited in private mode until publication, if supported by the corresponding repository. We will also implement these mechanisms for repositories that only allow submission through a website. Depending on the type and volume of data, we will also support an optional direct submission of the data to a public repository. It is additionally planned to automatically propose suitable repositories for data deposition based on the provided metadata.

Besides submitting data to public repositories, another important part of data handling is the ability to access and search for data in public repositories. We will develop a mechanism to find and load data from public repositories. Users will be able to perform queries across multiple repositories simultaneously. Frequently accessed repositories and datasets can be manually or automatically cached inside our storage infrastructure to avoid unnecessary repeated transfer of the same datasets. We will combine our efforts with NFDI4Biodiversity, who are planning a similar mechanism, to develop this system, if the requirements match.

Certain types of datasets, especially those that contain highly sensitive data, require special care before submission to external repositories. Thus, we will collaborate closely with NFDI4Health and, in particular, GHGA to make these datasets accessible under appropriate conditions of data security and privacy. This will include assisting researchers in handling these datasets correctly prior to submission.

The implementation of these features will be based on an extensible architecture. The wrapper implementations for the individual repositories will follow object-oriented programming design principles to build a collection of separate modules (repository adapters), which are embedded into the platform. This modular approach allows us to easily integrate modules that are developed in other communities as well, since this consortium alone will probably not have sufficient resources to implement adapters for all relevant public repositories. During the initial development, we will select 2-3 highly relevant repositories to showcase and test the service components and to collect user feedback as quickly as possible. Table 25

Contributors: ZB MED (lead, 12 PM), UFZ (co-lead, 5 PM), DSMZ (in-kind)

Goals: To establish digital preservation solutions for data types that do not have dedicated repositories. This includes enriching data with metadata, developing standards, raising awareness, and providing training and consulting on this topic for community members.

Providing digital preservation services: NFDI4Microbiota will provide the microbiology community with (1) bitstream preservation of (meta)data in the de.NBI cloud (M4.2 - Data storage platform), (2) preparation of files (e.g. formatting) for digital preservation, and (3) digital preservation of (3.1) molecular data in EMBL-EBI repositories and of (3.2) other data types in ZB MED’s dark archive through publication in the PUBLISSO Repository for Life Sciences (FRL) *38. The FRL enables authors to publish and archive life sciences research data in a way that makes them permanently accessible and citable. This data may be linked to a publication or self-contained (i.e. raw research data). Research data published in the FRL must be licensed as open data (i.e. opening up the possibility of subsequent reuse) and provided with a detailed description and metadata to ensure it can be clearly interpreted and reused in the future. ZB MED intends to make it possible to publish open-source software on the FRL. The FRL is based on Rosetta, a scalable digital preservation system built upon the OAIS reference model and relevant metadata standards such as Dublin Core, PREMIS and METS (M2.1 - Data & metadata standards). Rosetta is run as a “dark archive“ to protect the data against unauthorized access. The digital resources available in the FRL can be searched via ZB MED’s LIVIVO discovery service *36. The specific search functionality for objects from the microbiological community will be implemented in the NFDI4Microbiota Hub (M3.1 - Central web portal).

Fostering standardization and quality assurance: To standardize and control the quality of archived (meta)data, NFDI4Microbiota will select and implement relevant metadata standards (e.g. PREMIS). Furthermore, it will select suitable file formats for digital preservation (M3.5 - Data deposition & repositories) and advise data producers and repository operators on file format and integrity checks (e.g. format identification, validation, checksums). Additionally, the consortium will foster collaboration between research data producers and other repository operators on topics such as technology and community monitoring, and standardization.

Awareness building and training: To make digital preservation an indispensable part of microbiological research projects, NFDI4Microbiota will train researchers in the digital preservation of research data and associated metadata, for instance during Research Data Management workshops (already established by ZB MED, M1.1 - Training and education). Furthermore, the consortium will partner with the research community to help define criteria for assessing the archival value of research datasets. We will also offer individual support to researchers (M1.2 - Support) to assist them with specific issues related to digital preservation. Table 26

Contributors: UFZ (lead, 12 PM), DSMZ (co-lead, sig. in-kind)

Goals: To create and promote an online platform for community development of NFDI4Microbiota systems using reviews and comments from users.

Reviews and comments are the medium that connects developers (in this case the NFDI4Microbiota consortium) and users. Reviews and comments are also the way users communicate with each other. In the NFDI4Microbiota network, they will be used not only to ensure that the different NFDI4Microbiota systems are connected to their users, but also as a form of input that allows the community to participate in the development of the various systems. Measure M3.7 will provide an online platform for the NFDI4Microbiota network that will enable users to review and comment on all the different systems for data. The online reviewing and commenting platform will be built on the basis of five key principles: 1) Build a connection between NFDI4Microbiota users and developers. We will use our reviewing/commenting platform to encourage users to ask questions and give suggestions. By doing so, we will give users the chance to comment freely on the different systems they use. This will also create a direct link between users and the developers of the different systems and help to make the systems more successful. 2) Create connections among NFDI4Microbiota users. We will create a platform where users can answer questions raised by other users. This will add a community-driven dimension to the development of the NFDI4Microbiota systems. 3) Develop a medium for improving user interaction. User discussions will be searchable, providing a long-term platform for community development and forging connections between users. 4) Create backlinks. Backlinks are links from one website to a page on another website. These play an important role in most search engine algorithms. The traffic created by reviews and comments will link the various NFDI4Microbiota systems to the keywords from the respective user reviews and comments. This will make NFDI4Microbiota visible not only to other NFDIs but also to anyone searching for keywords relevant to the NFDI4Microbiota Network. 5) Users can rank the relevance of reviews and comments from other users. This will be used to help NFDI4Microbiota system developers rate the relevance of questions and comments and prioritize changes and new developments in the system. In other words, reviews and comments will be used as tools to make the NFDI4Microbiota system more community-driven. Table 27

Contributors: EMBL (lead, 46.5 PM), UFZ (co-lead, 23 PM), ZB MED (sig. in-kind)

Goals: To make it easier for users to contextualize their data and extend the scope and depth of their research through facilitated integration with complementary data.

Data integration can greatly enhance the value of individual datasets by increasing the magnitude and diversity of sampled populations and the biological facets they measure. The typical workflow for molecular profiling of microbial cultures or communities involves quantitative screening of one or more data types – such as 16S RNA profiling, (meta)genomics, (meta)transcriptomics, (meta)proteomics, and metabolomics – to provide a systems-level view of microbiota and microbial communities, enable meta-analyses, and inform modeling of biological processes. NFDI4Microbiota will facilitate integration between different data types within a study (i.e. the same set of samples measured in different ways) and between data from different studies (different sets of samples, measured in the same or different ways). Integration will have three goals: (1) to provide linkages from a user’s dataset to relevant public datasets and databases that add context and knowledge, (2) to make it easier for users to combine and compare their data with public data in order to add context and boost confidence in their results, and (3) to enhance the quality of the analyses that can be performed on a dataset and expand the biological questions that it can answer through multi-omics integration. This measure will design and implement integrative analyses and make them available to users through the central web portal and/or the analysis interface (M3.1 - Central web portal and M3.2 - Analytical services).

Integrative data linkage: User access to public complementary data and relevant biological models will be supported though integrative data linkage, which will help users discover connections between studies and with relevant databases. For example, genetic data from an in vitro microbial drug exposure experiment could be linked to metagenomic data from human fecal samples from patients who took these drugs, allowing the user to determine whether in vitro toxicity is reflected in decreased species abundance in situ. This will be possible thanks to the consistent way in which researchers upload their metadata connected to an experiment (M2.1 - Data & metadata standards). More generally, existing public data that are relevant to a user’s dataset will be computationally identified and gathered from the NFDI4Microbiota system (enabled by M3.1 - Central web portal and M3.3 - Databases & terminology services) and reported centrally to the user (enabled by M3.2 - Analytical services).

This integrative data interface will support inter- and intra-omic linkage between datasets based on the structured data generated from Task Areas 2 and 3 and the storage infrastructure in M4.2 - Data storage platform. Some of these linkages will then further support integrated analyses across datasets.

Integration of analysis results: When analyzing a dataset, users often want to compare their results to those of previous similar studies, or run a meta-analysis across studies. This can be challenging if the data from other studies are not easily accessible, the metadata is not properly linked to the data, or the analysis tools are computationally expensive or difficult to run. The NFDI4Microbiota platform provides a valuable opportunity to easily combine and compare results from many datasets. This is possible because the exact software versions and running parameters will be known for all results generated from analyses executed on the NFDI4Microbiota platform (supported in M3.2 - Analytical services). The results of consistent analyses run on public datasets can therefore be programmatically gathered and merged with a user’s results, providing valuable external context or validation. Datasets will be selectable thanks to their well-formatted and consistent metadata (M2.1 - Data & metadata standards, M3.5 - Data deposition & repositories).

Synergistic data analysis: The combined analysis of different profiling technologies – such as 16S RNA profiling, (meta)genomics, (meta)transcriptomics, (meta)proteomics, and metabolomics – can be synergistic and often adds confidence and value to molecular readouts. Combining different omics techniques can further provide a multifaceted view of a biological system and improve the creation of biological models of cellular processes in microbial species, as well as of their interactions with biotic and abiotic factors, the latter being fundamental for understanding microbial functionality and being predictive at an ecosystems level. Combining different types of data for microbial communities is, however, a very challenging task which requires a high level of expertise in, and access to, different technologies and is therefore lacking in many studies. This integration will thus be guided by benchmark studies and achieved through integrative analyses that will help microbial researchers to perform and develop cutting-edge multi-omics research. Multiple omics data types will be integrated using existing and newly developed tools and approaches. The biologically informed integration of different data types leads to new biological questions and thus improves the quality of research results.

By providing a unified, centralized database of diverse microbiological data with consistent storage, processing, and metadata formats, the NFDI4Microbiota platform creates the opportunity for powerful data reuse and integration. In this measure, this potential is harnessed to create synergy across datasets and experimental approaches and empower systems-level microbiological research. Table 28

Contributors: ZB MED (lead, 12 PM), RWTH (co-lead, 16 PM), FSU Jena (in-kind)

Goals: To foster the use of a common Electronic Lab Notebook (ELN) framework and create guidelines for selecting and using ELNs.

ELNs play an important role in documenting research data: they provide clear documentation of experiment planning and implementation and of data generation and processing. The aim of this measure is to ensure research data are captured as early in the process as possible in order to feed them directly into the analysis pipeline.

ELNs provide a digital means of organizing all the necessary metadata, which can then be used to create or update data management plans (DMPs). Over half of those who responded to our community questionnaire say that they intend to use, or are already using, an ELN solution. Additionally, the results indicated that more than 10% of respondents were not aware of ELNs as a potential solution. This shows a clear need to build awareness.

This measure seeks to cover two cases: firstly, situations where ELNs are already in use and, secondly, situations where help is required in choosing an appropriate tool. In the first case, various different ELN systems are available; we therefore intend to provide a range of interfaces to ensure compatibility, especially since there are currently no data exchange standards for ELN tools. In the second case, the selection of a suitable ELN can be aided by a needs analysis in the co-applicant’s or participant’s institution. It is important to remember that a variety of users will work with the ELN, including researchers, laboratory specialists, laboratory managers, and IT administrators. Each user will have their own specific requirements. Some requirements are common to all users, including the ability to map the most important lab-specific workflows and processes and fully document the research process. Criteria for the needs analysis will be based on the ELN Guide, which aims to help users select an appropriate ELN tool *39. Key criteria include usability, compliance with good scientific practice, the option of integrating the tool in institutional research data management workflows, IT security, and costs. The output of a needs analysis mainly comprises answers to the following questions: Is there a need for one or more ELNs? Do existing commercial or open source ELN tools match the users’ needs, or is it necessary to develop a more personalized solution?

Since ELNs can be used in just about any field of research, we intend to work on this topic in close cooperation with other consortia. By pooling resources with partners in the life sciences and beyond (e.g. NFDI4Chem), we can take the development of open source software to the next level. The key is to compile the needs of the microbiology research community in a clearly structured way. Table 29

Fig. 12

Contributions: The majority of work in this task area evolves around the provisioning of services for end users. All co-applicants are involved in at least one of the individual measures. Major leading roles are taken by ZB MED, HZI, DSMZ, UMR, JLU, UFZ, and EMBL. Each co-applicant will contribute their broad knowledge to the development of the various components and services and apply their experience to creating user-friendly applications. This includes competence and experience on a technical level as well as domain-specific knowledge in the various domains of microbial research.

Dependencies: The services and workflows implemented in TA3 will utilize the infrastructure provided by TA4. This includes use of the various platforms developed in M4.3 for the provisioning of DBMS as well as workflow execution backends such as SLURM or Spark clusters for more advanced scalable data analysis. It also includes use of the storage platform developed in M4.2 for storing data objects along with their associated metadata. In addition, the individual components will be monitored using the tooling developed in M4.4. All metadata handling, which is a cross-cutting topic within all parts of TA3, is based on the metadata standards defined by M2.1 and the metadata validation system developed in M4.2. The workflow standards from M2.3 are required for M3.2 and M3.4 to implement a proper provenance service.

Risks of implementation and mitigation approaches: In the event that the planned coordination between the measures of TA3 and TA4 does not prove to be efficient enough, the involvement of software architects will be increased in each of the relevant measures. Moreover, in the event of any inconsistencies or a failure to adopt standards from M2.1 and M2.3, we will boost the overlap and exchange of personnel and agree on more detailed requirement specifications. Finally, if the usability of the offered platforms turns out to be sub-optimal, we will use the feedback from user surveys to tailor the platforms to community needs.

Contributors: BIBI (lead, 30 PM), EMBL (co-lead, 18 PM), JLU (2 PM), RWTH (sig. in-kind)

Goals: To provide scalable cloud compute resources for the NFDI4Microbiota community.

As a national research data infrastructure, NFDI4Microbiota will offer compute and storage resources to its community members. Currently, the “standard model of computation” in life sciences requires users to first download and install software repositories locally before transferring large amounts of data from public databases. Cloud computing enables economies of scale by pooling compute and storage resources and offering a model where public data is hosted by data providers, allowing users to perform their analyses close to where the data resides. Virtual environments allow maximal flexibility in terms of software stacks, and portable containers enable scientists to share environments or workflows with collaborators and guarantee reproducible research.

In collaboration with the de.NBI Cloud, we will deploy services and data in scalable cloud computing environments. The de.NBI Cloud, part of the German Network for Bioinformatics Infrastructure (de.NBI), is a German academic federated cloud computing infrastructure for data analysis in the life sciences. As such it provides compute (>30,000 cores) and storage (>40 PB) resources including reference data and relevant bioinformatics methods. Coordination and governance of the de.NBI Cloud take place at Bielefeld University. Together with JLU and EMBL, three of the seven de.NBI Cloud sites are actively participating in the NFDI4Microbiota consortium. As part of ELIXIR-DE, we are already well integrated in ELIXIR and EOSC-Life (Demonstrator D.3, Metagenomics) and will also collaborate closely with other NFDIs such as GHGA and NFDI4Biodiversity.

M4.1 - Computational infrastructure operations will establish, coordinate and operate the technical infrastructure and offer the capacity and expertise to host the computational infrastructure for NFDI4Microbiota. This includes delivering cloud infrastructure, allocating resources, optimizing performance and capacity, ensuring proper resource management, maintaining availability in order to satisfy the needs and expectations of the community, and meeting service level agreements. de.NBI operates a cloud governance board that supervises the fair, reasonable, and economical usage of cloud resources. Together with M4.4 - Service monitoring & reporting, we will adopt and extend a similar governance structure according to the needs of NFDI4Microbiota. Access to resources will be ensured by a uniform authentication and authorization infrastructure (AAI), developed and maintained by ELIXIR, which is already running very successfully in the de.NBI network. Collaboration is foreseen with other NFDIs from different domains (in this case PUNCH4NFDI) to explore the benefits of establishing an NFDI-wide AAI. Depending on community needs, resources will be provided on different service levels, such as Infrastructure as a Service (IaaS), Platform as a Service (PaaS) and, in later stages, even serverless computing services. Table 30

Contributors: JLU (lead, 86 PM), BIBI (co-lead, 20 PM), FSU Jena (6 PM), DSMZ (sig. in-kind), UFZ (sig. in-kind), RWTH (sig. in-kind)

Goals: To develop and provide a scalable and easily accessible domain-agnostic data storage platform.

The core data storage system for NFDI4Microbiota will provide a domain-agnostic, extensible, and scalable platform to store the data of the various user communities and allow fast, efficient and secure access. Essentially, the object storage provided by the de.NBI cloud will be used to store the actual data objects (files). As the requirements and technologies used for this measure are highly similar to the infrastructure to be developed within the NFDI4Biodiversity consortium, and due to several overlaps among co-applicants, an agreement has been made to develop the data storage platform jointly. We have therefore planned additional features and several advanced extensions that will be implemented primarily within NFDI4Microbiota. By complementing NFDI4Biodiversity’s software development plans and combining our forces, we will be able to create a versatile and powerful distributed data storage platform with a comprehensive and tailored set of features, including high availability, optimized georedundant storage of datasets, local and distributed execution of analysis workflows, and automated versioning. Data items will be stored inside the object storage provided by the de.NBI cloud, and we will use a document storage database such as MongoDB for the accompanying metadata. A group of objects will compose a dataset, and a subset of the objects of a dataset can be tagged as a dataset version. Objects can also be aggregated to object groups that act as directories. Objects and dataset versions will be tagged according to semantic versioning principles (e.g. v2.5.27-3.alpha).

The platform will be accessible via standardized and well-documented REST APIs that are provided by automatically scalable Kubernetes deployments. To make optimal use of the scalability provided by the object storage back-end, the actual upload and download of objects, although negotiated via the REST APIs, will be carried out directly via the S3 interface of the object storage. Primarily, the REST APIs will be designed as low-level APIs to be used by software developers who create applications and services for end users such as those described in TA 5. The metadata will be expected in JSON format.

Since the NFDI4Biodiversity consortium will have commenced work on the basic functionality of the storage platform in 2021, it will be possible to provide an early alpha version of the storage platform at the start of the NFDI4Microbiota development phase. Although this early version will be far from feature complete, this should allow for quick and easy adoption of the features that are already implemented, as well as early testing.

The storage platform will consist of two components: (1) a local storage module to handle and manage data storage, and (2) a distributed component to connect locally deployed components. During the implementation of the distributed system, development of the local storage module will continue and new features will be added based on user feedback. We have reached an agreement with the PUNCH4NFDI consortium to continuously share our ideas and experiences relating to cloud storage architectures on a technical level. Initially, we will collaborate on the design of a technical solution to implement tailored caching mechanisms for data and metadata (e.g. based on XCache) and evaluate further possibilities for cooperation. Table 31

Contributors: JLU (lead, 67 PM), BIBI (co-lead, 15 PM), EMBL (4 PM), UMR (13 PM), UFZ (6 PM), RWTH (sig. in-kind)

Goals: To provide ready-to-use computational frameworks for data analysis and computing services to the NFDI4Microbiota community.

This measure will provide a set of common frameworks to support computational analyses and the development of data processing services such as command-line tools, analysis workflows, and webbased bioinformatics software applications. All frameworks are either required by other measures as IT infrastructure components or are utilized directly as tools by users of the NFDI4Microbiota platform (i.e. usually software developers) to facilitate common computational tasks as well as the implementation and hosting of bioinformatics software tools. We will also provide some general tools for software developers and services to be used for reproducible data analysis, e.g. Juypter notebooks.

Most of these tools will be configurable via our NFDI’s online hub to make them easily accessible for non-technical users. The planned tools and services can roughly be divided into three groups: (1) core infrastructure components, (2) database management systems, and (3) tools to support the development of analysis workflows and software applications.

The core infrastructure components will be offered as Platform as a service (PaaS) solutions and will be configurable via the NFDI4Microbiota web portal. Many workflow engines still rely on traditional HPC environments for distributing large compute tasks. Parallelization is implicitly defined by the input and output specifications of the individual processes. The resulting workflows are inherently parallel and can be scaled up or scaled out according to the volume of the data to be processed. By using job queuing systems like SGE or SLURM, many computational tasks can be executed in parallel. Based on e.g. the BiBiGrid framework developed at BIBI, we will support simplified and highly automated setups of scalable HPC compute clusters in cloud environments, including an Ansible-based deployment of queuing systems (SLURM) and other analytical frameworks. As a core part of the technical infrastructure, this measure will also deploy and operate a set of Kubernetes clusters to host the services of this consortium to allow fast and scalable deployments.

The second type of platforms are database management systems (DBMS) for various purposes. We aim to support a wide range of database types including traditional relational DBMS such as MySQL and PostgreSQL as well as NoSQL databases such as MongoDB, Apache Cassandra, Neo4j and Redis.

As a third goal, we will make access to common data analysis tools easier and help users during the life-cycle of data analysis and software development (M3.2 - Analytical services). Thus, existing software tools that are popular in the user community or those that are requested by participants will be made available in the cloud (e.g. by using container-based deployment), which will facilitate their usage for analyses described in M3.2 - Analytical services. In addition, we will offer analysis notebooks such as Jupyter and Zeppelin as a service.

Within this measure we will also provide a set of tools and interfaces to support stream-based data processing. This will enable direct connections between our central storage platform and local datageneration devices, such as Oxford Nanopore sequencing instruments that are used in larger sequencing centers, smaller laboratories and even in the field. Based on edge computing technologies, client software can be used for local data preprocessing prior to data upload (e.g. to reduce the volume of data). As the preprocessing is use-case dependent, the required preprocessing tools for different analytical workflows will be developed in M3.2 - Analytical services. The upload can also be initiated automatically on the client side; subsequently, further automated and pre-configured data processing steps can be triggered based on the callback system of our storage platform. Table 32

Contributors: BIBI (lead, 30.5 PM), UFZ (co-lead, 12 PM), all other co-applicants with sign. in-kind

Goals: To identify appropriate indicators to monitor the quality and acceptance of NFDI4Microbiota’sservices and repositories, to establish a data-oriented service index, and to measure the impact of ourjoint efforts.

Service monitoring – i.e. the continuous recording, processing, and reporting of a defined set of technical and user-based measurements – is nowadays quasi-standard for any productive operating service. It supports a multitude of administrative, developmental and reporting functions during the lifetime of a given service, ultimately increasing its quality, usability and visibility. Key to the whole NFDI initiative is the idea of maximizing community involvement in order to connect data producers with data and service providers and eventually with data and service users. To this end, NFDI4Microbiota will identify and establish service performance indicators on different abstraction levels such that they (i) monitor service life and usage status, (ii) allow services to be adopted and developed according to community-defined needs, (iii) increase the acceptance of FAIR data policies and leverage the usage of FAIR services and repositories, and (iv) provide a measure of the impact of NFDI4Microbiota on national and international communities.

Monitor life status and usage statistics: An initial measure in this task area will be to establish a Service Control Board (SCB) that brings together expertise from all service levels. Its main purpose will be to make sure that agreements on service-related issues are properly addressed in the task areas. The SCB will be open to all consortia members but will initially consist of the leads of M3.2 - Analytical services, M3.4 - Data quality & provenance services, and M4.1 - Computational infrastructure operations, M4.2 - Data storage platform, plus the lead and co-lead of M4.4 - Service monitoring & reporting. The SCB’s first task will be to develop a service guideline that outlines a set of minimal requirements such that each service offered by this consortia meets a minimum standard of service quality. This guideline is further complemented by general usage and quality metrics tailored to different service categories.

Develop and adapt to community-defined needs: In addition to the overarching exchange with the microbiota community addressed by the Board for Community Outreach (BOC, M1.3 - Community outreach and public relations), it is vital to gather user feedback on a per-service basis in order to identify current shortcomings and future needs. To address this task, we will work closely with the SCB to establish a user feedback form for all analytical, data quality, and data storage services as well as the central NFDI4Microbiota web hub. Feedback will be fully transparent to service developers and providers, allowing them to respond to the community in a timely, direct and interactive manner.

Increase acceptance of FAIR data policies: One of the major challenges for all NFDI consortia is to inspire the scientific community to produce, process, and share data in a FAIR manner. The SCB’s first job will be to address this challenge by monitoring data usage statistics as an integral part of the data and metadata repositories (as described in M3.4 - Data quality & provenance services and M3.5 - Data deposition & repositories) and linking this back to the service monitoring framework. By doing so, data producers are rewarded by increased usage statistics as a form of a data credit system and, moreover, actively connected to data usage instances that can lay the foundation of future collaborations.

Measure impact and success of NFDI4Microbiota: The SCB will develop and maintain a platform for all NFDI4Microbiota partners and co-applicants and for project governance. This platform will aggregate and integrate all the statistics of the aforementioned measurements and tasks into a single entity with full transparency, making it possible to quantitatively and qualitatively assess the impact of NFDI and NFDI4Microbiota on the research communities. Table 33

Fig. 14

Contributions: As TA4 is the backbone of the whole consortium with various closely connected objectives, all contributors within this task area will likewise work in close collaboration and reach out to all the other co-applicants and participants that will actually use the infrastructure. Specifically, BIBI and EMBL will coordinate all efforts regarding cloud infrastructure operations across the different cloud sites, additionally supported by their partner JLU. JLU and BIBI are responsible for the deployment of the data storage platform and will primarily contribute to the storage API and the storage authentication model, supported by DSMZ, UFZ, and FSU Jena. Accompanied by BIBI, JLU will also lead the development of software tools and common components supported by EMBL, UFZ and UMR. This includes generalpurpose platform services and frameworks to facilitate the development, utilization and deployment of analytical frameworks and tools. The definition of appropriate service indicators for monitoring and reporting, their implementation on all service levels, and the development of a monitoring framework will all be coordinated and carried out by BIBI and UFZ.

Dependencies: TA4 will support the integration into the NFDI4Microbiota platform of all software components developed in the measures of this task area and will connect the corresponding services with M5.2 for easier monitoring and reporting.

Risks of implementation and mitigation approaches: (1) Potentially unreliable storage services will be addressed by redundancy of software components and using multiple cloud sites; (2) limited usability based on difficulties in integrating the provided system will be prevented by allocating additional software architect PMs in measures that directly interact with the infrastructure components; (3) disagreement with other NFDI consortia on joint development of platform components will be avoided by defining a minimum set of common platform features and joint adjustments in prioritization of planned developments.

Contributors: ZB MED (lead, 24 PM), HZI (co-lead, 24 PM)

Goals: To provide a functional governance structure for this consortium, to establish the consortium’s internal bodies, and to bring the voice of the community into the consortium.

The NFDI4Microbiota Administration Unit (AU), supported by ZB MED and HZI administrative staff, will be established to meet these goals. Together with the Board of Directors (BoD), it will coordinate the establishment and work of the internal consortium bodies. A preliminary Administration Unit, consisting of Alice C. McHardy, Konrad U. Förstner and their project management staff, has already been established to coordinate the writing and submission of this proposal. The Administration Unit will coordinate exchanges between consortium (co-)applicants, as well as other NFDI consortia, the SAB, user council and user community (represented by participants). The Administration Unit will work closely with the BoD to monitor scientific activities within NFDI4Microbiota and release an annual report. This measure also includes drawing up the collaboration agreement between consortium members and suitable by-laws, building on the experience ZB MED gained from drawing up similar documents for NFDI4Health. It will also generate openly accessible documentation of the operations and processes to lay the foundations for clarity and transparency for all stakeholders. To allow for successful and synergistic collaboration between partners across the various TAs, several task groups will be established. Preliminary versions of these have already been formed to work on parts of this proposal. Additional task groups will subsequently be established by the Administration Unit if necessary. Finally, the Administration Unit will explore different options for establishing an NFDI4Microbiota extranet. The most suitable option will be implemented as part of Task area 4 - Technical Infrastructure. This platform will support internal consortium processes and facilitate project management, document and data sharing, and enhanced collaboration between consortium partners. Table 34

Contributors: ZB MED (lead, 36 PM), HZI (co-lead, 36 PM)

Goals: To monitor the financial plan for all tasks, organize possible future extensions of the consortium, and report within the consortium as well as to funders.

ZB MED will set up project financial controlling and serve as the central contact point to provide advice on financial matters, supported by HZI. It will prepare management-level justification of resources deployed and the summary financial report. The Board of Directors is responsible for financial reporting to the DFG. The legal basis of financial controlling and reporting will be laid out in a cooperation agreement which will be signed before the funding period. Annual project costs will be calculated and forecast once a year, in September. If necessary, an application for shifting funds to the following year will be filed. The financial requirements of co-applicants for the coming quarter will be collected by ZB MED on a quarterly basis. ZB MED will request the funds from the DFG and distribute the funds to co-applicants without delay. A financial overview and justification of spending for the previous year will be collected by the ZB MED annually in mid-February and communicated to the funder in March.

For internal reporting, progress within task areas is monitored by the TA leaders and reported by them on a regular basis (semi-annually) to the Board of Directors and the General Assembly. The BoD reports to the SAB. External reporting duties are the responsibility of, and coordinated by, the BoD. The BoD will keep the funder (DFG) and the NFDI Directorate informed of any changes and developments in the project.

Financial plans for including further participants will be drawn up by the BoD and decided upon by the GA. Any necessary revisions of the financial plan will be prepared by the BoD with support from the Administration Unit and decided upon by the GA. In addition, the BoD will organize financial planning and all transactions for internal NFDI4Microbiota project calls.

A mid-term report will be drawn up after three years.

Table 35

Contributors: HZI (lead, 6 PM), UMR (co-lead, 5 PM), all other co-applicants with in-kinds, ZB MED (285 PM)

Goals: To continuously adapt our strategy and services to meet community needs, offer state-of-theart methodologies and follow best practices, while establishing and maintaining the NFDI4Microbiota infrastructure.

Technology exploration and development of new use cases: The consortium will draw on its close ties with its community to develop new use cases (M1.3 - Community outreach and public relations). As well as gathering user feedback, we will also take into account the recommendations of the NFDI4Microbiota experts themselves and the NFDI4Microbiota SAB. Existing use cases will be adapted accordingly and new ones developed. Concepts for the required analytical services will be updated accordingly (M3.2 - Analytical services). In addition, the consortium members, being experts in their research areas and well-connected in the community, will periodically assess the state of novel technologies for infrastructure development and emerging data types and decide where the framework can feasibly and sensibly be adapted and extended.

Flexible funding and allocation: To realize the dynamic adaptation of services and support of further data types, develop new training programs, and improve our ability to react to unforeseen technical challenges, we will establish a flexible fund allocation mechanism for a substantial portion (∼10%) of the total funding. These so-called Flexible Funds Projects will be awarded annually through a project call process with external peer review. For this purpose, a Project Review Board (PRB) – an external body whose members will be selected based on suggestions made by the SAB – will be implemented. It will consist of renowned scientists in changing compositions, based on the topics required for review and excluding members with potential conflicts of interest. The PRB will (1) be convened as needed to review proposals for innovation and implementation projects, and (2) provide an unbiased review of innovation and implementation projects. Based on these reviews, the General Assembly (Organizational structure and viability) will select projects that best serve the overall mission of NFDI4Microbiota.

Growth: We will also explore further incentives for data provision, such as reimbursing data holders for the upload of their data, for instance, depending on the data quality, size, provided metadata, and need for further curation. In addition, we will foster the exchange of data between different platforms by interacting with other NFDI consortia, such as GHGA, NFDI4Biodiversity and NFDI4Health (M1.4 - Connection to other NFDI consortia), as well as international culture collections, highthroughput culturing initiatives (Maier et al. 2018), and microbiome platforms, such as MGnify, MG-RAST, IMG Integrated Microbial Genomes and Metagenomes, and large data-generating consortia such as the Human Microbiome Project (I and II), Tara Oceans and Global Top Soil.

Table 36

Fig. 16

Contributions: The aim of TA5 is to provide the organizational and administrative foundations for all other TAs. To this end, ZB MED and HZI – represented by K. Förstner and A. McHardy – will establish the required structures. They will be supported in this task by the Administration Unit. HZI and UMR will jointly work on the adaption of future developments with support from JLU.

Dependencies: The activities of TA5 depend on the cooperation of all other TAs in terms of providing information and reports which are then compiled in overall reports. Additionally, there are numerous external dependencies in the interaction with the DFG and the NFDI directorate.

Risks of implementation and mitigation approaches: (1) Future development might require adaptations of activities and their budgets which could lead to conflicts. These will be addressed by direct communication of the BoD with the affected co-applicant and might be escalated to the GA to make decisions. (2) In the event that parties do not deliver the agreed deliverables in the required quality or time, other parties might take over and funds will be reallocated.