In 2012, the Digital Universe Study reported that ‘less than 1% of the World’s data is analyzed and less than 20% is protected’ and recommended that management plans were needed to harness the potential within these data streams (Gantz and Reinsel, 2012). Life science is a data science that is dependent on the generation, sharing and integrative analysis of vast quantities of digital data. However, these data are complex and fragmented, creating a significant barrier to their integration and reuse. Data are generated for different research purposes at thousands of facilities across the world, with diverse formats, annotations, methodologies and metadata standards. Interpreting this data is difficult and often involves integrating very large datasets from multiple sources. Certain data types such as sensitive human data archived under controlled access require protection, so data federation and sharing are a challenge in this context (Saunders et al., 2019). Since its official launch in December 2013, ELIXIR (https://elixir-europe.org/), the European infrastructure for life science data, has worked to address these challenges by bringing Europe’s national centers and core bioinformatics resources into a single, coordinated infrastructure. Many European countries have long-established national bioinformatics infrastructures that provide tools, data resources and cloud services to national and international users. ELIXIR invests in and aims to sustain these vital resources and ensure a level of interoperability that facilitates scientific discovery. Indeed, for research to thrive in a world of data abundance, all components (research data, analysis tools, standards as well as computational resources and training materials) must be FAIR—findable, accessible, interoperable and reusable (Wilkinson et al., 2016). ELIXIR supports European life scientists in making data FAIR, ensuring that the benefits extend far beyond those actively participating in ELIXIR, and, additionally, encourages all funders to adopt Open Science mandates. Here, we describe how ELIXIR has evolved to be an established research infrastructure, with a review of significant milestones achieved, and a look into the future. By linking national networks that span most of Europe’s major research centers ELIXIR is in a unique position to drive the transformation of life science’s distributed data resources within Europe to be sustainable, federated, standards-based and cost-effective. ELIXIR is now a mature intergovernmental organization that has grown from 6 members in 2014 to 22 members and one observer in 2020. ELIXIR links more than 220 institutes and over 700 national experts dedicated to the development and operation of national services that contribute to data access, integration, training and analysis for the research community (see Fig. 1). Overview of the ELIXIR infrastructure including members, services, infrastructure, training and industry events Fundamentally, ELIXIR is connecting national bioinformatic networks, known within ELIXIR as Nodes (https://elixir-europe.org/about-us/who-we-are/nodes). Some ELIXIR Nodes predate ELIXIR, such as EMBL-EBI and the Swiss Institute of Bioinformatics. In many other countries, ELIXIR has contributed to the formation of new national bioinformatics infrastructures, e.g. de.NBI (https://www.denbi.de; ELIXIR Germany) and ELIXIR Czech Republic (https://www.elixir-czech.cz/), both launched in 2015. The participating countries contribute membership fees, proportional to the countries Net National Income (NNI), provide ELIXIR with a technical budget that is used to fund development of shared services that support and link the nationally funded bioinformatics resources. The Nodes guide the direction of ELIXIR in alignment with their own national priorities, as has been described, e.g. by ELIXIR Netherlands (Eijssen et al., 2015), ELIXIR Switzerland (Baillie Gerritsen et al., 2019) and ELIXIR Norway (Tekle et al., 2018). ELIXIR aims to ensure that all life science projects, irrespective of their membership status with respect to ELIXIR, have access to advanced, scalable and long-term sustainable infrastructure across Europe. In this expanding interdisciplinary field, training of new researchers and infrastructure providers is key for success and ELIXIR runs a vibrant training network. For example, between September 2015 and March 2019, ELIXIR was able to register over 1200 training materials and train more than 19 000 people. Capacity building of this magnitude depends upon the coordinated reuse of skills, resources and materials. Partnerships between industry and academia are increasingly important if the societal impact of scientific research is to be fully realized. For instance, access to open data and software has been shown to be of long-term value to the bioeconomy (Bousfield et al., 2016; Rothe et al., 2019). Industrial users of ELIXIR’s services range from large multinationals to micro-Small and Medium Enterprises (SMEs), and include the pharmaceutical, healthcare, food, agriculture and blue biotech sectors. ELIXIR has shown that many SMEs extensively rely on life science data from public resources (Garcia et al., 2018). ELIXIR aims to build new partnerships and strengthen existing collaborations (Lauer et al., 2019) through its innovation and SME Forum, which has delivered 10 events over the first program, covering topics as diverse as Genomics and Health and Data-driven innovation in the agri-food industries. ELIXIR facilitates collaboration between its member institutes and researchers with two intersecting organizational groupings, the Platforms (https://elixir-europe.org/platforms) and the Communities (https://elixir-europe.org/communities). The ELIXIR Platforms (Data, Compute, Interoperability, Tools and Training) are supported by Technical Coordinators located at the ELIXIR Headquarters, with vision and strategy led by senior scientists in the Nodes. As part of joining ELIXIR, each Node contributes a set of services (https://elixir-europe.org/services) that are aligned with the Platforms, e.g. Human Protein Atlas (Uhlén et al., 2015) (Data Platform—ELIXIR Sweden), CSC Cloud (https://research.csc.fi/) (Compute Platform—ELIXIR Finland), Identifiers.org (Juty et al., 2012) (Interoperability Platform—EMBL-EBI), CAMEO (Haas et al., 2018) (Tools Platform—ELIXIR Switzerland) and TeSS Training Portal (Beard et al., 2020) (Training Platform—ELIXIR UK). The Node-contributed services form the basis for ELIXIR development work. Established and driven by researchers and bioinformaticians in the Nodes, the ELIXIR Communities identify the needs of domain- or technology-specific research around a specific theme, e.g. Metabolomics, Microbial Biotechnology. ELIXIR now has a total of eleven Communities (see Fig. 2) which is a significant progression from the four use cases (Federated Human Data, Marine Metagenomics, Plant Sciences and Rare Disease) that were supported by the initial Horizon 2020 ELIXIR-EXCELERATE grant. Harrow et al. (2021) describe how ELIXIR can support the research communities in their effective use of life science data, as exemplified by the EXCELERATE use cases. ELIXIR’s life science research Communities and Platforms are linked by Implementation Studies To ensure engagement between Nodes, Platforms and Communities, ELIXIR conducts Implementation Studies (https://elixir-europe.org/about-us/implementation-studies) that develop services and bring benefit to key user communities (see e.g. Fig. 2). To date, more than 50 Implementation Studies, generally lasting between 6 months and 2 years, have been initiated. These are deliberately of short duration, timely, focused on bottlenecks highlighted by a Community or Platform, and their completion generally provides analysis components, workflows or validations. They are selected based on high scientific or technical merit and broad applicability, typically addressing different stages of the data life cycle, e.g. secure data deposition, distributed data annotation or advanced data reuse. They drive development of the service portfolio and generate opportunities for building bundles of services bespoke for particular communities. ELIXIR Communities and Platforms illustrate how ELIXIR fosters long-term collaborations that connect national resources into transnational infrastructure. The federated data infrastructure is held together by strong community standards, with users and data providers agreeing on conventions for annotating, depositing, finding and accessing data. ELIXIR’s distributed organization with a scientific leadership that represents 23 national bioinformatics Nodes is a powerful vehicle to consult and build consensus on joint standards and drive the implementation in national organizations and communities. For instance, the ELIXIR Plant Sciences Community has built a common technical infrastructure and associated social practices to support plant genotype-phenotype analysis with the goal to make plant genotypic and phenotypic data easier to find, integrate and analyze, by making them FAIR (Pommier et al., 2019). The Community developed and proposed an extension of the Minimal Information about Plant Phenotyping Experiments (MIAPPE) v1.0 specification (Krajewski et al., 2015) and is recognized as a significant player in the global plant community and is working on MIAPPE with EMPHASIS (https://emphasis.plant-phenotyping.eu/), the ESFRI for European Plant Phenotyping, and the CGIAR (https://www.cgiar.org/), the world’s largest global agricultural research organization. It is particularly beneficial in rapidly developing fields to bring together leading experts across national centers to collaboratively develop infrastructure and standards. ELIXIR’s Marine Metagenomics Community has addressed a lack of accessible databases by providing curated reference data that, together with benchmarking datasets and standardization of computational pipelines, provides the necessary infrastructure to enhance both academic and industrial research and development. New databases such as MarRef, MarDB, MarCat and METdb provide important new reference data sources for interpretation and species assignment within the wider marine metagenomics community (Robertsen et al., 2017). A software ‘container’ packages code with all its dependencies, enabling the application to run quickly and reliably across a range of computing environments, facilitating reproducible research. The BioContainers Implementation Study is an example of an initiative driven by the ELIXIR Tools Platform to provide a stable infrastructure for unifying software containerized solutions within ELIXIR. This infrastructure provides an access point for end-users to find, generate, store, monitor and even benchmark software containers. This has resulted in the development of a registry for containers, BioContainers (da Veiga Leprevost et al., 2017), which allows researchers open access to 7.9 K containerized tools. The Bioschemas convention, set up in 2015 to improve FAIRness of data, is driven by the ELIXIR Interoperability Platform as an open international community initiative. Bioschemas improves the findability of data in the life science resources by extending a Schema.org markup to specific biological data types. This markup can be introduced into websites, without the need for specific technical expertise, such that they become readily indexable by search engines and other services (Profiti et al., 2018). ELIXIR continues to encourage the development of Bioschemas through its yearly Biohackathon (https://www.biohackathon-europe.org/), which brings developers together from different Platforms and Communities to collaborate on preselected projects. ELIXIR also funds strategic infrastructure projects such as AAI, the Authorization and Authentication Infrastructure (https://elixir-europe.org/services/compute/aai). AAI facilitates data access that requires specific authorization on account of its sensitive nature, e.g. human genomic data. Before data access authorization and delivery can be carried out, a researcher needs to be identified and their identity authenticated to a sufficient level of assurance. AAI provides this function. ELIXIR AAI is still in development, but already has 603 enabled identity providers and 2470 registered users organized in 423 groups, with 59 production services connected to it. Genomic technologies have advanced such that generating human genomic sequence data is no longer prohibited by cost and time. It is projected that by 2025, over 60 million patients will have their genome sequenced in a healthcare context (Birney et al., 2017). Infrastructure for secure access to sensitive human data is not currently developed and there is often little opportunity for the reuse of data outside, or beyond, the core project. ELIXIR is driving the strategy toward sustainable and long-term infrastructure for sensitive human data in Europe (Saunders et al., 2019). For example, the ELIXIR Federated Human Data Community develops solutions to overcome the access difficulties for data resources containing genomics and linked phenotypic and clinical data that are posed by general data protection regulations (GDPR) policies (Phillips, 2018). The ELIXIR Rare Diseases Community (https://elixir-europe.org/communities/rare-diseases), in partnership with RD-CONNECT (http://rd-connect.eu/), BBMRI-ERIC (http://bbmri-eric.eu/) and E-Rare (http://www.erare.eu/), is creating a federated infrastructure that will enable researchers to discover, access and analyze different rare disease repositories across Europe. This community is using Beacon (Fiume et al., 2019), a genetic variation discovery tool initially developed by the Global Alliance for Genomics and Health (GA4GH), to share variant information from consented sensitive human genetic data stored in databases affiliated with ELIXIR Nodes and in the European Genome-phenome Archive (EGA). GA4GH is an international, nonprofit alliance formed in 2013 to accelerate the potential of research and medicine to advance human health (https://www.ga4gh.org/about-us/). ELIXIR has been a formal contributor to the generation of GA4GH products (https://www.ga4gh.org/genomic-data-toolkit/) since the formation of the alliance. The ELIXIR Beacon project was one of the first GA4GH Driver Projects (Fiume et al., 2019). In January 2017, to further develop and implement the Beacon technology across ELIXIR Nodes, the ELIXIR Beacon project adopted new ELIXIR AAI features such as tiered access and improved security, to minimize risk around individual privacy. The project has over 16 000 users accessing the Beacon API and to date more than 70 Beacons have been lit. In May 2019, a Strategic Partnership between ELIXIR and the GA4GH was announced (https://elixir-europe.org/news/elixir-and-ga4gh-expand-collaboration), building on the existing collaboration, and will facilitate the access to sensitive data across Europe, in order to help create virtual cohorts with tens of millions of participants. The coordinated development and deployment of the ELIXIR Beacon and ELIXIR AAI standards along with the suite of standards from the GA4GH Work Streams will help translate this vision into reality. ELIXIR’s mission includes ensuring that all of Europe’s life science projects have access to long-term sustainable infrastructure for data management (Martin et al., 2019). The five strategic objectives of the ELIXIR 2019–2023 Scientific Programme (https://elixir-europe.org/about-us/what-we-do/elixir-programme) include that ‘ELIXIR Core Data Resources will be the global standard for bioinformatics resource management and the foundation for an international funding and life cycle management strategy that secures the long-term sustainability of those resources’. Defined according to a protocol based on 23 indicators described in the article ‘Identifying ELIXIR Core Data Resources’ (Durinx et al., 2017), the ELIXIR Core Data Resources (https://elixir-europe.org/platforms/data/core-data-resources) are a set of European data resources of fundamental importance to the wider life science community and the long-term preservation of biological data. Having this ELIXIR has an analysis to of and open data and FAIR data integration, and funding for these data resources et al., This is the first such and, ELIXIR is participating in the Global et al., 2017), an initiative an international of funders and and working toward long-term sustainability of Core Data Resources has been to potential funding et al., 2018) and is that this challenge can be addressed in international long-term sustainability needs to extend beyond core data this has enabled on the to life science bioinformatics service and training particularly as we a of of genomic health and The long-term sustainability of the life science data depends also on the of formats, and across data resources. ELIXIR has established a portfolio of Interoperability Resources to facilitate interoperability of life science data, to support the of FAIR data and to enable and of life science data, in the The first ELIXIR Scientific Programme was supported by ELIXIR Core funding and the Horizon 2020 ELIXIR-EXCELERATE Programme of Horizon enabling the from a to an established research infrastructure. As shown through the ELIXIR has the foundation for a sustainable life science that its 2019–2023 Scientific Programme The 2019–2023 Programme ELIXIR from the to at This transformation of is in Implementation Studies have two or Nodes with for ELIXIR now as a distributed European infrastructure in multiple For example, the Federated Human Data Implementation Study in Nodes organized into five and will run for The includes a for ELIXIR Nodes to the European Genome-phenome Archive federated by providing the necessary and training across the network. This will enable phenotypic and data to be accessible across international and provide a foundation for major such as the by European countries to provide access to at million human from national by (Saunders et al., 2019). federated with open tool and standards will enable analysis workflows to on data across national Open repositories and support for development will drive data and which are of high importance in both research and clinical The ELIXIR Implementation Study and Cloud the need for a stable infrastructure for unifying software solutions within ELIXIR. This infrastructure will provide an access point for end-users to find, generate, store, monitor and even benchmark software drive the technical experts in ELIXIR’s and Tools Platforms will with developers and researchers in the and In both of these the of and as well as the for generating reusable infrastructure across diverse communities is of the over the of ELIXIR links 23 Nodes and on national to develop and services that enable data access, integration, analysis and training for the research ELIXIR has been recognized as a Infrastructure of Global by the and brings together and resources from more than 220 A key of ELIXIR’s strategy is to build long-term around services identified as community ELIXIR’s on connecting and value to the national Nodes a portfolio of shared services a foundation for long-term sustainable in national research infrastructure This with other such as the to et al., and a common infrastructure was the funding was by support for the necessary the of the initial ELIXIR has upon its Scientific Programme which the scientific direction between ELIXIR Nodes. ELIXIR has been an for data sharing and This project will Node across ELIXIR, and ELIXIR’s vital in driving data and reuse in a distributed and funding In the ELIXIR’s in large infrastructure provides the basis to translate genomics from research into application in healthcare ELIXIR Nodes are in large genomic health projects such as European Programme on Rare Diseases and Infrastructure for National in Europe, and and ensuring wider of Node services and within the health ELIXIR is also the FAIR data and of data from selected projects funded by the and data from industry In plant and agriculture research a data federation Europe’s largest plant centers is now fully (Pommier et al., 2019). This provides the foundation of a European federation of data repositories dedicated to which will the of distributed plant ELIXIR’s in the Global that the and et al., developed by the ELIXIR Data Platform to identify Core Data Resources and are recognized by the wider international ELIXIR’s in driving Open as a for funded research. ELIXIR is the project the ESFRI life in Europe to create an open digital for life science in the The project aims to data from the life Infrastructure facilities and centers as FAIR Data Resources in the link reusable Tools and to services in national life science connect all users across Europe to a AAI and to develop joint data policies to and the by research and patients their data and ELIXIR has collaborations with a of international such as GA4GH (see and the Data In is new collaboration with national data infrastructures in key international countries such as the et al., 2019) around and infrastructure, so that both will benefit and their ELIXIR is also now extending engagement with the wider international community on such as vision is to support a of to the complex between and genetic and disease and across In ELIXIR is established as a sustainable infrastructure for and large and distributed datasets based on global standards and shared components that value to national The from ELIXIR’s member countries, as described in the ELIXIR 2019–2023 is to provide services that enable European researchers and their to access, analyze and reuse complex and distributed the years, ELIXIR will the existing partnerships and open up collaborations with new such as and and to drive the vision to support life science research and its to the and The are to the contributed to the around this and in particular we to and for help the and for on the They also and for of the and on how to improve for the general ELIXIR has funding under the European Programme of Horizon 2020 ELIXIR-EXCELERATE of

SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

"Scientific community" refers to a group of people collaborating together on scientific-research-related activities who also share common goals, interests, and values. Such communities play a key role in many bioinformatics activities. Communities may be linked to a specific location or institute, or involve people working at many different institutions and locations. Education and training is typically an important component of these communities, providing a valuable context in which to develop skills and expertise, while also strengthening links and relationships within the community. Scientific communities facilitate: (i) the exchange and development of ideas and expertise; (ii) career development; (iii) coordinated funding activities; (iv) interactions and engagement with professionals from other fields; and (v) other activities beneficial to individual participants, communities, and the scientific field as a whole. It is thus beneficial at many different levels to understand the general features of successful, high-impact bioinformatics communities; how individual participants can contribute to the success of these communities; and the role of education and training within these communities. We present here a quick guide to building and maintaining a successful, high-impact bioinformatics community, along with an overview of the general benefits of participating in such communities. This article grew out of contributions made by organizers, presenters, panelists, and other participants of the ISMB/ECCB 2013 workshop "The 'How To Guide' for Establishing a Successful Bioinformatics Network" at the 21st Annual International Conference on Intelligent Systems for Molecular Biology (ISMB) and the 12th European Conference on Computational Biology (ECCB).

We consider a class of weighted nuclear norm optimization problems with important applications in signal processing, system identification, and model order reduction. The nuclear norm is commonly used as a convex heuristic for matrix rank constraints. Our objective is to minimize a quadratic cost subject to a nuclear norm constraint on a linear function of the decision variables, where the trade-off between the fit and the constraint is governed by a regularization parameter. The main contribution is an algorithm to determine the so-called approximate regularization path, which is the optimal solution up to a given error tolerance as a function of the regularization parameter. The advantage is that we only have to solve the optimization problem for a fixed number of values of the regularization parameter, with guaranteed error tolerance. The algorithm is exemplified on a weighted Hankel matrix model order reduction problem.