The genomes of thousands of Dutch individuals are now profiled annually in healthcare and research. This has greatly increased diagnostic yield, enabled to choose the best medicine with least adverse effects, has provided many clinically actionable insights and has dramatically increased our understanding of genotype-phenotype relations. Still, 60-70% of the rare disease patients is left without a diagnosis, choosing the right (expensive) advanced cancer treatment for patients who do not respond well to conventional treatment is still more or less comparable to flipping a coin, and routine use of genomic data to enable personalized medicine is still an unfulfilled promise.
Integrated analysis of all these genomic data holds the promise to dramatically improve healthcare benefits and research progress. However, the valuable genomic data is fragmented and hard to access across many healthcare providers and research organizations, prohibiting extraction of the potential genotype-disease knowledge. Therefore this project aims to promote large scale (re)use of all human genomic data in the Netherlands to maximize knowledge extraction for research and healthcare, while considering needs of all stakeholder groups including patients/participants and addressing ELSI issues. This aligns well with the ambition of The Netherlands to actively participate in the Million European Genomes Alliance (MEGA) .
Hereto this project will invest in harmonization of essential aspects of the genomics data workflows and reporting across all data-generating centers in the Netherlands. Based on the ZonMw hiatus analysis we will in particular bridge barriers in Findability, Accessibility, Interoperability and Reusability (FAIR) aspects of genomics data, including bundling of all relevant genomic and phenotypic metadata necessary for knowledge extraction, process metadata for quality and health technology assessment (HTA), and a ‘shrink wrap’ license to protect patient/participant rights and consents. The project will therefore deliver a broadly supported unifying guideline that gives practical solutions to increase FAIRness of genomics [2,3]. The scope of this guideline is limited to next generation DNA sequencing data (at least Whole Genome Sequencing (WGS), Whole Exome Sequencing (WES) and panel analysis applicable to germ line and somatic analyses) but we expect to expand into RNA, proteins, metabolites and the microbiome soon following this project, e.g., in the framework of X-omics / BBMRI / ELIXIR / European Joint Programme for Rare Diseases.
Specifically, this guideline will consist of 1) a white paper on requirements and best practices, with reference to international solutions (GA4GH, Solve-RD/ESHG/Eurogentest, EGA, TCGA, ICGC); 2) a data and metadata standard recommendation to formally and unambiguously capture the (meta)data needs of all stakeholders (research and clinical setting, including phenotypes and HTA data); 3) implementation of findability of genomic data by expanding on the Dutch Health-RI catalogue 4) provisioning of best practice protocols and solutions for data sharing and access including innovative privacy-friendly approaches to give access to information that is encoded in the data without actually having to share the sensitive data itself (e.g., Personal Health Train, Beacons, Federated access, aggregated data, VWdata). We believe it essential to involve data generation centers from the start to ensure the guideline is practical and does not incur unwanted extra cost. To test this we will therefore 5) include report of independent implementation studies in ZonMw GGG projects, complemented by X-Omics, Hartwig nationally, and Solve-RD in collaboration with ESHG/Eurogentest internationally (all committed to contribute financially and/or intellectually to this project) which then can be used as reference. Finally, we will 6) promote uptake and sustainable maintenance of the guideline by embedding the guideline within the relevant professional groups, healthcare and research organisations (all in support of this proposal), coordinated by Health-RI and GO-FAIR.
We will measure immediate project impact by uptake of the guideline into the best practice guidelines and standard operating procedures of health-care professionals (VKGL, VKGN, PALGA, NVVP, NVHG working groups, the hospitals/labs) and large research infrastructures (X-Omics, BBMRI-NL, DTL, ELIXIR-NL, ELSI service desk). In addition, Health-RI will monitor long-term impact measurement of increase of findability (e.g., via Health-RI catalogue), measurement of access statistics (all partners), and use of interoperability recommendations and reuse (via citations of data). Ultimately, beyond the lifetime of the project we expect increased diagnostic and healthcare data (re)use, and the Dutch research community back to the forefront of genomics research (relative to other countries having invested in massive centralized genomics projects).