The Department of Health and Human Services, National Institutes of Health has a requirement for Secure Platforms Support for the N3C Data Enclave.

The mission of the National Center for Advancing Translational Sciences (NCATS) is to transform the translational science process in order to get more treatments to more patients more rapidly. Critical to this mission is the ability to manage data, generate insights, and foster collaboration across National Institutes of Health (NIH) Institutes and Centers and with external partners such as research hospitals, academic institutions, other government agencies, and industry stakeholders. NIH, NCATS, and NCATS collaborators are among the world’s leading researchers, and they have access to large, cutting-edge data sources. An increasingly complex challenge is to translate these disparate, ever-evolving data sets into actionable insights that accelerate the pace of science and clinical development. Tools are needed to enable teams to ask complex, multi-faceted questions and to collaborate more seamlessly and securely across various teams.

NCATS has established the NCATS Secure Scientific Platform Environment, a specialized cloud-based data aggregation and analytics enclave that can integrate, manage, secure, and analyze any kind of scientific data, and provide secure, controlled access to internal and external collaborators.

NCATS requires a secure cloud platform-as-a-service (PaaS) that can support the National Clinical Cohort Collaborative (N3C) Data Enclave. The NCATS Secure Scientific Platform Environment (the “Environment”) is a specialized cloud-based data aggregation and analytics enclave that can integrate, manage, secure, and analyze any kind of scientific data, and provide secure, controlled access to internal and external collaborators. Within the Environment, multiple NIH institutes and centers (ICs), Federal agencies, and Federal task forces integrate, manage, secure, and analyze all types of scientific data using dedicated platforms, and, equally importantly, make that data available in specific and controlled collaborations with each other and with external collaborators.

Within the Environment, multiple NIH ICs, Federal agencies, and Federal task forces integrate, manage, secure, and analyze all types of scientific data using dedicated platforms, and, equally importantly, make that data available in specific and controlled collaborations with each other and with external collaborators.

The Environment is a mission-critical data management and analysis environment for multiple data management efforts by several NIH ICs and Federal agencies under the leadership of NCATS. The Environment currently uses Palantir Technologies, Inc’s Foundry platform as the incumbent contractor.

This platform-as-a-service (PaaS) has supported NCATS, the National Cancer Institute (NCI), the President’s Emergency Plan for AIDS Relief (PEPFAR), and the National Clinical Cohort Collaborative (N3C). The Environment has integrated hundreds of live intramural and third-party data sources in support of dozens of ongoing, critical scientific projects that rely on continuous access, data, and analyses within the Platform. The Environment is now the standard means for accessing and collaboratively analyzing NCATS screening data for dozens of investigators at both NCATS and NCI, and for accessing and analyzing RNASeq and several kinds of proteomics data at NCI. The Platform has also supported clinical applications supporting NCATS (the Clinical and Translational Science Awards (CTSA) and Rare Diseases Clinical Research Network (RDCRN)), NCI, and PEPFAR.

The Contract shall provide the secure platform-as-a-service, software licenses, professional services, and cloud hosting to support the N3C Data Enclave as follows:

• Professional Services:
  • Host the N3C Data Enclave in AWS GovCloud
  • The current N3C vendor has achieved FedRAMP High Authorization for its Platform as a Service (PaaS) cloud product. The N3C Data Enclave does not require the High Impact Level, but the platform does require, at minimum, a FedRAMP Moderate Authorized PaaS offering
  • Configure the Environment’s data ingestion infrastructure to support N3C’s requirement to ingest, join, clean, and harmonize clinical data from all participating clinical sites. Clinical data from sites will be ingested, harmonized, and structured to significantly increase the scale of the research data asset
  • Integrate additional medical and clinical ontologies and other datasets. This will enable N3C researchers to build a more comprehensive knowledge base and link it to their scientific research to enable novel tagging and discovery workflows and provide the potential to explore new hypotheses. The Contractor will also configure additional live data connections to build upon the foundational data asset, as requested by NIH and the research community. This will enable EHR data to be linked to other types of data, such as pathology, genomics, and social determinants
  • Enable research projects on top of N3C through configuration of analytical templates, toolkits, packages, and workflows. The Contractor will continue to work closely with N3C researchers to collaboratively scope and configure additional research workflows within the N3C Platform. By integrating data from multiple sites, the platform will enable researchers to explore questions with vastly more statistical power than is achievable at individual clinical sites, which is essential for better understanding
  • Configure the Environment’s machine learning and artificial intelligence (AI) framework to enable more advanced research and analysis of large-scale clinical data. The Environment must be configured to perform production-grade machine learning analyses—including using graphical processing units (GPUs)—on clinical data. These capabilities enable more advanced research maximize the usefulness of a centralized approach to analytics. The Environment must support full model lifecycle management, including building, tuning, retraining, evaluating, and monitoring
  • Train and onboard N3C researchers on collaborative analytics tooling within N3C. In a “train the trainer” approach, the contractor, in collaboration with NCATS, will continue to train and onboard extramural researchers in both the US and other approved locations, such as the European Union (EU), to fulfill the vision of a truly shared, collaborative research infrastructure centered around the NIH and its extramural programs
• Software Requirements:
  • A commercial software solution deployable on day one of the project that can be configured within expedited timelines
  • An open data architecture, where data always remains under the full control of NCATS and other data owners and can be easily exported in open, non-proprietary data formats via open APIs. The software should be built on an open, distributed microservices architecture with well-documented REST APIs and out-of-the-box connectors that are designed to seamlessly interface with other systems, adapt to meet evolving needs, and avoid system lock-in
  • Proven multi-modal data integration capabilities, including the ability to rapidly ingest electronic health/medical record (EHR/EMR) data (including OMOP, TriNetX, ACT, PCORnet, etc.), pathology samples and assay data, unprocessed high-throughput drug screening (HTS) outputs, genomic data (including bulk RNA-seq, scRNA-seq, CITEseq, TCRseq, ChIPseq, Microarray, etc.), imaging data (e.g., MRIs), mass spectrometry, flow cytometry, and other data types used in basic and translational biosciences research and public health, such as administrative, financial, and grants data (e.g., nVision, IMPAC II, I2E, myDCEG, ARS, NIDB, PubMed), and supply chain data. Backed by configurable and interoperable data quality checks and a Git repository for data pipelining
  • A multi-tenant secure enclave backed by configurable governance and access policies. Ability to host multiple individual tenants, with subsets of data shareable with different parties in the model as desired and in accordance with access controls. Access to a single view of multi-modal data based on user group and/or role
  • Proven granular security controls with the ability for data owners to control all downstream uses of the originating data easily and dynamically, and the ability to conform to NCATS security policies. Ability to request and grant selective access to levels of data sensitivity in-platform based on a user’s intended purpose, implement configurable governance workflows depending on the requirements of data use and data transfer agreements, and audit user behavior after access has been provisioned
  • The ability to maintain data and scientific provenance and reproducibility of all integrated data sources. Every resource (dataset, analysis, code, plot, report) contains provenance, metadata, and can be both traced back to the exact version of all upstream dependencies, and where the dependency tree can be easily replayed given new data or updated analysis logic, while still retaining prior versions and branches
  • Dynamic data model, object-based search/discoverability, and analysis workflows, allowing easy definition of objects, properties, and links that propagate from a source table. Solution provides natural ways to move between tabular and object-oriented interfaces and data analyses. Proven ability to integrate multi-data model data into a harmonized data model (such as OMOP)
  • Intuitive, highly configurable user interfaces that have been effectively configured and utilized by technical bioinformaticians, cheminformaticians, data engineers, and data scientists, as well as less technical biologists, chemists, clinicians, analysts, program managers, administrators, and other users.
  • Ability to perform advanced analytics and informatics (including management of machine learning and other models) in a user’s preferred open coding language, as well as in point and click tools, all within the same environment. Ability to generate no-code analytical templates enabling less technical users to conduct complex analyses and generate visualizations
  • A variety of proven, secure, and user-oriented configurable applications and workflows backed by configurable access controls and up-to-date data, including:
    • Patient digital twin capability for tens of thousands of patients backed by multi-modal data (e.g., clinical, imaging, and tumor sequencing/mutation data)
    • Laboratory sample and result tracking system
    • Streamlined research funding analysis, tracking, and reporting interface for improved funding estimates and budget oversight.
    • Genomic pipeline code templates for generating analysis and publication-ready visualizations
    • Application for sharing and re-use of research outputs. A centralized space where logic, datasets, models, and other research outputs can be securely shared, discovered, and re-used by other researchers. Usage of each artifact should automatically be tracked to ensure attribution for contributing researchers. The application should ensure that use of shared artifacts is compliant with governance rules around data use
    • Application for creation and management of code sets. This should allow automatic integration and updates for multiple terminologies and include the ability to version code sets, track their usage, and document them with metadata such as their provenance and intention. Changes to vocabularies should be tracked and users should be alerted when these changes impact existing code sets.