Best Practices for Implementing Informatics Systems for R&D Collaborations

Posted on Lab Informatics. 15 July, 2019

In today’s global economy, scientific organizations in many different industries are turning to collaboration with external partners to fuel their R&D pipelines with flexible networks of researchers. These external collaborations can take many forms – research institutes, industry and academic partners, contract research organizations (CROs), contract development and manufacturing organizations (CDMOs), external testing laboratories, consortia, etc.

Many organizations combine numerous partners in diverse ways across multiple research projects. Even in simpler models, any collaboration with an external partner is typically not static, but evolves over time. Therefore, sponsoring organizations are often changing the business processes around the collaboration frequently and rapidly.

While external collaboration can provide many benefits including improved flexibility, enhanced innovation, and reduced time-to-market, externalized R&D activity introduces unique data and IT challenges. Some of these include:

  • Synchronization of partner and in-house data across different transactional systems
  • Maintaining secure and appropriately partitioned data
  • Harmonizing master data to facilitate high quality data flow
  • Developing appropriate long-term plans for data management, including potential data repatriation in an efficient manner
  • Protecting intellectual property (IP) and managing joint IP

These challenges can result in additional costs and be potential limitations on the benefits of external collaborations. At the least, they introduce risks for sponsoring organizations. All too often, unfortunately, these data and information management aspects of a collaboration are not fully considered until problems arise. In this blog, we discuss key characteristics of informatics systems for collaborations, along with best practices for implementing a collaboration platform.

Approaches to R&D Collaboration Data Management

Nearly all R&D informatics systems are designed and implemented only to meet internal R&D requirements. Also, organic growth of internal R&D activities often leads to a tangled web of processes and systems with significant assumptions incorporated into the ecosystem. These latent aspects of systems frequently become impactful when considering the flow of data outside of the R&D organization, and how to open internal systems and/or their data to external collaborators. Some examples of system characteristics important in collaborative data flows are:

  • User identity and access management processes and technology
  • Data access control models
  • Processes that require multiple systems with human-only integrations (“sneakernets”)

Sometimes these limitations make it infeasible to use the internal systems and processes with an external collaborator. Although it may seem more efficient to design our systems with external collaborations in mind, the reality of delivering informatics capabilities on budget and in time almost always means this does not happen. When faced with supporting external collaborations, this leaves the following choices:

  • Use the collaborator’s system. If the collaborator is in the business of collaborations, they are potentially more likely to have systems that would meet the challenges above.
  • Transfer data in email attachments. This lowest common denominator approach and unfortunately tends to be the status quo.
  • Implement a new informatics capability.

There are important sub-aspects to the implementation of a new capability. First is the relationship to the existing system(s). If the current system is meeting requirements and is only insufficient for use in a collaboration, then considering how the system might be extended is an appropriate course of action. If the current system is lacking, or there’s a likelihood of long-term multiple collaborations, then a strategic assessment with the development of a roadmap to a solution architecture that meets all needs is essential.

If either a significant extension of an existing system or a new system is needed, then a cloud-first solution architecture should be considered. Cloud-first systems have several distinct qualities that make them a logical choice to meet the needs for R&D collaboration data management. Specifically, these qualities are:

  • Configurable by intent
  • Based on a tenant model for data and configuration
  • Built for automated deployment

Key Characteristics of a Cloud Collaboration Platform

Some important characteristics of potential candidates for a cloud-based R&D collaboration data management solution are:

Configuration. An ideal platform is highly configurable, allowing organizations to define sites, projects and user roles and, along with the access permissions for each. The user authorization mechanism should be able to incorporate company-specific identity and directory systems for ease of use by scientists, rather than having separate identify and password management for the collaboration system. The platform should also support a range of core R&D capabilities, potentially including some of the functionality of:

  • A flexible, multi-disciplinary ELN
  • A portal that allows scientists for sharing reports, protocols, and other documents
  • Data capture, analysis and visualization capabilities

Deployment. An effective cloud-based platform allows quick creation of separate collaboration environments for use with specific partners. Each environment should represent a secure data  partition. Data from an environment should be extractable and be able to be merged into other environments.

Security. Cloud providers should have ISO accreditation for their systems, technology, processes, and data centers. Data should be encrypted at rest and during transit using well-defined current best practice encryption techniques. The collaboration platform data architecture should have strong isolation across tenants and include logging of all system access and use.

Integration. The cloud platform should have a complete and robust programming interface (API) for integration with the internal systems of either organization in the collaboration. The platform must support bi-directional integration and data syncing between on-premise systems and cloud applications.

Best Practices for Implementing a Cloud Collaboration Platform

There are several best practices that should be followed to successfully implement an effective cloud collaboration platform. These include:

Strategic Planning. One of the most important steps in successfully implementing a cloud collaboration platform is the planning necessary to ensure project success. Towards this end, the first steps in the project should include a thorough workflow and business analysis in order to develop the optimized future-state requirements that guide the technology selection process. In addition, an end-state solution architecture should be developed, along with a strategic roadmap to deployment. Good strategic planning helps ensure the deployment effectively and efficiently meets business and technical needs.

Change Management. It is important to carefully consider the cultural impact, employee training, and new policies that necessary to ensure success of the new collaborative environment. Since collaborative R&D informatics systems by definition involve employees of multiple organizations, attention to change management for these systems is of paramount importance to success.

Efficient Testing. Bandwidth requirements for cloud computing are significant, and load and volume testing are important to ensure that system performs acceptably. Waiting until late in the project to discover that your system is not capable of handling the data transport requirements causes unnecessary scrambling to meet implementation goals.

Effective Validation. As some vendors claim prevalidation for their cloud-based software, it is important to understand exactly the scope of the Install/Operational/Performance Qualifications this covers. Compliance requirements mandate validation in the user’s environment, so prevalidation does not suffice to fully satisfy regulations. Working with the vendor to clearly establish individual and joint responsibilities for validation prevents unnecessary duplication and establishes an overall credible approach.

A Detailed SLA. Working with the vendor to create a detailed SLA is one of the most important things you can do to ensure a successful implementation. Without a well-written SLA, your organization could be in for many unpleasant surprises and additional expenses down the road. In addition to system change management processes and requirements to maintain compliance, an important and often overlooked aspect of the SLA is data storage, including controls of underlying data replication related to availability and disaster recovery.

Conclusion

In today’s increasingly collaborative R&D landscape, creating and managing informatics systems to help scientists handle, analyze and share information is critical for organizations to enhance innovation and remain competitive. Cloud- based collaborative platforms can provide a secure, scalable and flexible approach for handling the wide array of data types, sources and partnerships which are involved in modern collaborative research. These systems allow organizations to spin up robust collaborative environments easily with minimal IT support.

When implemented properly, cloud-based research informatics systems as a complement to R&D collaborations can provide important benefits to your organization:

  • Effective use of data produced from the collaboration
  • Increased scientist productivity
  • Enhanced organizational flexibility and agility
  • Reduced IT costs per user

There are attractive benefits to a cloud-based collaboration research informatics system, but implementation of the platform can be a difficult endeavor that requires much skill and planning to execute successfully. The project team should follow a proven, comprehensive methodology in order to ensure that the implementation provides significant business value for your organization.

Astrix Technology Group has over two decades of experience in the laboratory informatics domain. Our professionals bring together the technical, strategic and content knowledge necessary to help you efficiently select, configure, implement and validate a cloud collaboration platform that best meets your needs and keeps your total cost of ownership low. Whether your deployment utilizes public, private, or a hybrid cloud architecture, our experienced and skilled professionals can make the process of implementing or migrating to a cloud collaboration platform far more cost effective and efficient. Contact us today for more information on leveraging the cloud to improve agility, reduce cost and advance collaboration when working on new scientific discoveries and technological innovation.

About Dave Dorsett

Dave Dorsett Dave Dorsett has more than three decades of experience in R&D informatics throughout the global pharmaceutical, chemical, and consumer goods industries. He has an extensive track record architecting, designing, and delivering commercial and in-house informatics solutions across the R&D spectrum, from early research through late-stage development.

A Selection of Current Customers