(Fast-Track proposals will not be accepted. Phase II information is provided only for informational purposes to assist Phase I offerors with their long-term strategic planning.) Number of anticipated awards: 1 to 3 Budget (total costs, per award): Phase I: $325,000 for 9 months; Phase II: $2,000,000 for 2 years It is strongly suggested that proposals adhere to the above budget amounts and project periods. Proposals with budgets exceeding the above amounts and project periods may not be funded. Summary: The unprecedented 2020 COVID-19 pandemic and subsequent physical distancing has led critical experimentation for developing novel diagnostic techniques and potential therapeutic interventions to nearly grind to a halt. This is due to the reliance on people to be physically present in a laboratory. The contract proposed here is to develop a next generation therapeutic discovery or diagnostic platform consisting of distributed, Artificial Intelligence (AI)-enabled, fully automated pieces of instrumentation that are capable of functioning in an autonomous fashion and directly linked to virtual cloud-based resources. The type of instrumentation required will depend upon the kind of experiment to be performed. Regardless of the experiment type, the instrumentation should be geared towards therapeutic discovery or a diagnostic technique such as an ELISA assay, the resultant experimental data could be ingested in real time to a cloud-based virtual research organization (VRO). The VRO model proposed would greatly improve safety with little or no loss of productivity. A key challenge for rapid development of diagnostics and therapeutics has been the inability to acquire, harmonize, store, analyze and share data generated during experimentation. Through this concept, a platform would be created to make data accessible to researchers anywhere on the globe in near real-time to help respond to a fast-changing pandemic or other healthcare crisis where time is of the essence. Project Goals: The goals of this project are to develop a platform comprised of three core components: 1. Distributed, modular, next generation autonomous laboratories that focus on areas such as high throughput screening (HTS) for drug discovery, next generation sequencing (NGS), high content imaging (HCI), polymerase chain reaction (PCR) diagnostics and others. 2. A cloud based VRO that each distributed automated laboratory is directly connected to. 3. Federated AI, potentially most critical, is the integration of the physical laboratories with the virtual cloud environment such that AI methods can be utilized to generate hypotheses based on previous experiments that could then be tested in the physical laboratory environment. This distributed and iterative approach would allow for the on demand initiation of a physical experiment, which could conceivably be a HTS run in one location along with NGS at another, the generation and analysis of data, with each experiment performed further expanding the available data to be used for more efficient and accurate AI models which can then initiate new experiments. It would be possible to compile and quickly perform a relational analysis of multiple relevant data types ingested into the cloud-based VRO from different distributed autonomous laboratories, as well as use AI generated hypotheses to trigger new experiments in order to broaden the dimension of discovery in a shorter time frame. The modular nature of this approach will also allow the entire platform to quickly add or scale up additional resources as required to respond to emerging pathogens or other biological scientific needs. Phase I Activities and Expected Deliverables: • Develop a prototype Platform for Rapidly Deployable Autonomous Laboratory comprised of three components: o Modular instrumentation with the following characteristics: • Must be used in typical laboratory operations such as HTS, NGS, HCI, PCR, etc. • Must use standard laboratory instrumentation communication protocols to communicate with other devices such as RS-232, TCP/IP, CAN bus, etc. • Must use ethernet based protocols to communicate with cloud-based environments such as TCP/IP, MQ Telemetry Transport (MQTT); commonly used for the Internet of Things (IoT); Advanced Message Queuing Protocol (AMQP), etc. • Must have a comprehensive application program interface (API) allowing for full control of the instrument, ranging from initiating the execution of an instrument specific protocol, to monitoring the status of the device, reporting results, reporting device faults with ability to recover, etc. • Must be able to communicate with other pieces of instrumentation to develop functional laboratory platforms. • Must include an instrument which generates data as the result from laboratory operations such as HTS, NGS, HCI, PCR, etc. • Must include consideration for a sample transport device. • Ideally will utilize modular components which are easily replaced when required. o A cloud based VRO that each distributed automated laboratory is directly connected to with the following core capabilities: • Infrastructure: • A scalable, cloud-based architecture residing in a major cloud service provider such as Amazon Web Services (AWS), Microsoft Azure, Google Cloud Platform (GCP), etc. The offeror should be able to demonstrate access (via an executed license) to the cloud service. • Data Storage, Access, and Catalog • Distributed data storage via services such as S3, Azure and Google storage etc. • Storage Class memory technology to be able to access data in real-time. • Data access API for simplified interface such as the use of 'PUT' and 'GET' requests. • A Data Catalog accessible via an API and keep tracks of all data sources and their respective metadata. • Data Extraction, Aggregation, Integration, and Harmonization: • Ability to integrate simultaneous data connections from multiple concurrent sources. • Enable connections to multiple types of databases and makes data available to users through a single access point. • Secure collaboration: • Multiple users should be able to visualize and work on the same data in a collaborative platform. • Remote Binding to ensure device control is handled securely and safely • Complaint with the following industry best-practice certifications, attestations, alignments, and frameworks such as: o SSAE18 SOC 2 Type II o ISAE 3000 SOC 2 Type II o FedRAMP Moderate (Ideally, but not required for Phase I) • Interoperability and Open Architecture • Store data in open source formats and expose REST APIs to interoperate with third-party and open source tools o Integration of the physical laboratories with the virtual cloud environment to allow for: • Federated AI and Machine Learning (ML) such that AI and ML methods can be utilized to generate hypotheses based on previous experiments that could then be tested in the physical laboratory environment. • Accessibility from external collaborator laboratories for use of the VRO and ability to remotely run experiments and process data into the VRO cloud environment • Adherence to appropriate safety protocols and procedures for physical control is mandatory of the above requirement within the VRO for anyone to be able to remotely control instrumentation. • Ability to scale, publish and share instrument services/functionality i.e. laboratory as a service (LaaS) • Provide cost estimates to develop a proof of concept platform capable of meeting the specifications listed above. • Provide NCATS with all data resulting from Phase I Activities and Deliverables. Phase II Activities and Expected Deliverables: • Build a prototype platform that meets the Phase I specifications. • Provide a test plan to evaluates all components of the platform, from the instrumentation performing some laboratory operation, to the data being sent securely to the VRO and finally with that data being used to propose and initiate a new experiment to be run on the platform. • Demonstrate that the platform is scalable to potentially hundreds of pieces of instrumentation in a distributed fashion. • Provide NCATS with all data resulting from Phase II Activities and Deliverables.