We present a novel informatics ecosystem that supports end-to-end FAIR data integration in clinical studies.  We use an example of biosample and biomarker processes to show how our ecosystem provides a flexible way to plan, define, ingest, quality check, store and share data from complex, largely outsourced clinical studies. We also show how we can enable efficient sample tracking, provide data analysis capabilities and offer a variety of scientific dashboards to provide a real-time insights of the status of ongoing studies.

The three most important aspects to consider when it comes to trying a new software product are: price, ease of implementation, and value-added. In this presentation, I will show how to convert every data visualization created in JavaScript, R, or Python into a complete and unobtrusive data analytics platform without the need for extra coding.

A core component of the CADD software strategy, Global Digital Manifold (GDM) is a high performing web-based application built around the concept of sharing a variety of best-in-breed computational chemistry tools without having to learn intricacies of each and avoid using multiple versions of shell scripts. GDM has allowed us to deploy a variety of tools such as generative chemistry (AI/ML), molecular dynamics, conformer generation, and docking. With some of these tools requiring more compute resources than others, GDM has been enabled to seamlessly switch between on-prem HPC to Cloud. In addition, this talk will also discuss a novel concept called "conveyors" available within GDM that allows flowgramming several tools together. These features have not only allowed better sharing, but higher adoption of in silico methods.

Numerous presentations are made on the need to bring down silos and make data accessible to foster a greater collaboration culture. This talk will share the various tooling and patterns put in place at pRED to address the ever-growing number of vendor and custom solution integration needs.

Creating high-fidelity digital twins of a patient’s vasculature provides a non-invasive means to measure blood flow conditions. Such models are gaining traction for determining the severity of stenosis and guiding treatment decisions. As we see an increased use of computational flow models for virtual surgery and treatment planning, there is an increased need to address two key challenges: providing an intuitive means for interaction and minimizing the computational burden. In our work, we are developing a framework for combining virtual reality interfaces with large-scale, parallel fluid mechanics modeling packages and using cloud-based resources. Virtual immersive environments can offer advantageous visualization of spatially heterogeneous and complex systems. However, as different VR devices offer varying levels of immersion, there remains a crucial lack of understanding regarding what level of immersion is best suited for interactions with patient-specific flow models. Here, I will talk about how we evaluate optimal immersion for different tasks and our experiences connecting the virtual reality software to a cloud-enabled backend for fluid simulation.

FAIR data handing is the basic requirement for successful data processing in the research area of the pharmaceutical industry. Effective ML/AI need Findable, Accessible, Interoperable and Reusable data to increase the data product within the research process. In terms of the administration of the steadily increasing number of chemical structures and their related data the normalization of the chemical drawings plays a central role in the data capturing of the research work. The growing requirements of the normalization systems makes it necessary to update the technical systems to modern highly efficient data management systems. To reach this goal Roche updated its structure normalization tools from the formerly used Cheshire to the BIOVIA Pipeline Pilot Chemical SDK using a cloud native technology stack. This talk illustrates the new IT architecture and reports the issues that have arisen during the implementation project.

Nine years ago, the FAIR principles were first discussed as basis for a FAIRPort, a FAIR Infrastructure for heterogenous data. A lot has happened in those nine years, and we ask ourselves the question: did we achieve machine-actionable data environments in this time? And if not, where should we be working on together to create FAIR infrastructures?

Research data portals, science gateways, data commons, and web applications need scalable, secure data management solutions. The Globus platform as a service enables application developers to provide robust file transfer, data sharing, search, and automation capabilities within their own life science research data applications and services, while leveraging advanced identity management, single sign-on, and authorization capabilities. The platform is built on widely adopted industry standards such as OAuth and OpenID Connect for authentication/authorization, and uses trusted protocols such as GridFTP and HTTPS. By outsourcing research data management to Globus, developers can focus their efforts on core application logic.

The Protein Sciences core labs in Genentech’s research departments engineer, produce, and manage large molecules for many scientific purposes. These core labs have to manage the many workflow processes at various stages and collaborate with each other and with the end users. A Protein and Antibody Request System (PARS) was created to simplify and coordinate molecule production processes in a centralized and collaborative tool.

This presentation will show how to make your software releases less riskier, while using Software Development Life Cycle (SDLC). Align development, testing and release management teams. Streamline the development and release process to meet delivery dates consistently while aligning the software objectives with business. Learn 1) the importance of acceptance criteria and how to make release when requirements are not well defined, 2) how to standardize the releases across your organization, 3) what minimum work needs to be done before releasing the software application, and 4) how to control releases.