Five Takeaways from Benchling's First User Forum
At Benchling, we're thrilled to be fostering a community of scientists and R&D managers. Tens of thousands of scientists - from the enterprise pharma companies to biotech startups to academia - use Benchling daily. To bring this community together, we hosted our first ever User Forum on October 18, 2018, in Cambridge, MA. It was a jam-packed afternoon of presentations and discussions, with some of the most forward-thinking R&D leaders presenting on their research, and how they use Benchling to optimize R&D. The speakers included:
Tiffany Chen, Director of Discovery, Rubius Therapeutics
Michael Dinsmore, Associate Director of Informatics, Editas Medicine
Scott Hamilton, Senior Lead Process Engineer, Synlogic
Christine Loh, Vice President Biology, Kymera Therapeutics
Dennis Underwood, VP Molecular and Information Systems, Agenus
Vipin Suri,Vice President Discovery, Obsidian Therapeutics
Throughout the presentations, and at the networking reception that followed, there was a real sense of excitement and optimism. We're at the start of a biotech revolution, and the R&D community is more ambitious than ever. Five common themes emerged during the presentations and discussions, revealing good practices for collaboration and data management in life sciences R&D:
1. Build R&D informatics with growth in mind
All of the presenters at the User Forum have seen considerable growth since adopting Benchling, which is a good problem to have. For Scott Hamilton, a Senior Lead Process Engineer at Synlogic, this meant doubling of the number of bench-top bioreactors in the lab, while adding two 24-reactor high-throughput machines and a real-time analytics machine.
One of the biggest challenges is processing the data as fast as, or faster than, we are creating it.
Scott Hamilton, Senior Lead Process Engineer, Synlogic
Scott deals with this challenge in two ways. Firstly, he uses Benchling's Workflow Management module to streamline data capture and minimize error-prone, manual tasks. Wherever possible, Scott leverages Benchling's open APIs to completely automate data capture and integrate results-generating instruments directly with the Benchling Data Warehouse.
Growing companies are just that: growing. R&D needs are evolving all the time, and often IT and informatics teams are still being built out. For Tiffany Chen, Director of Discovery at Rubius Therapeutics, growth means constantly developing new prototype cell therapies. When Rubius adopted Benchling, it was because unstructured data management approaches were no longer meeting the needs of this cutting-edge R&D organization. Rubius has used Benchling to standardize and centralize data, while remaining flexible. The ability to customize workflows through a point-and-click user interface is crucial. This allows R&D software at Rubius to evolve alongside R&D with minimal effort.
Benchling has helped a lot as we have grown as an organization, because the system is very configurable. If we have different groups with different priorities of how they want to organize their workflows, they can create their own.
2. To improve collaboration - move away from point solutions
Effective communication and collaboration are prerequisites for making better R&D decisions. When results and institutional knowledge are scattered across different systems, it makes it impossible to agree on a single source of truth. This challenge was touched on by several presenters when describing the state of data management prior to adopting Benchling.
Benchling has really put the data at our fingertips, whereas before it was so spread out and difficult to follow.
Benchling's platform approach to R&D informatics has helped customers like Rubius Therapeutics centralize data, making it easier to collaborate and share across teams. For example, prior to adopting Benchling, requests from service teams at Rubius were managed with meetings, emails and spreadsheet logs. Benchling's Request Management module allows scientists to reference exactly which registered entities they are requesting, which are linked to the workflows and protocols required to produce them. Teams can agree clearly on deliverables, and view progress in real time, avoiding ad-hoc communication and misunderstandings.
Benchling has helped a lot with our communication, because it's one site for all of our key information. It holds all of our sequence information, all of our data, and all of our experimental protocols.
3. Focus IT resources on the interesting problems
Michael Dinsmore, who leads informatics at Editas Medicine, spoke about using Benchling as a developer platform. Editas leverages Benchling's open APIs to implement time-saving custom solutions. Michael discussed recent projects, including data visualization, automating sequencing pipelines, and generating primer plates. Importantly, Editas uses Benchling's unified data infrastructure as a foundation for custom solutions, avoiding the need to waste IT resources integrating point solutions.
Benchling handles all the basics for us, and we're left to the interesting problems.
R&D informatics should provide a strong foundation on which to integrate equipment and implement custom solutions, without the need to start from scratch. Connecting disparate systems, one-by-one, in an attempt to centralize data, or building on-premises custom solutions, is extremely expensive and time-consuming. Inevitably, you end up with a system that doesn’t really work, breaks often, and is so difficult to use that scientists and managers will effectively ignore it.
4. Organize and structure your data for AI and advanced analytics
The panel discussion at the Benchling User Forum focused on R&D management in the current age of unprecedented technological change. Panelists were particularly interested in the promise of advanced analytical techniques, machine learning, and augmented intelligence, and their applications in drug discovery.
Agenus is really deeply into machine learning and AI. We see it as a necessary component of moving forward. But before you get to that point, you need to wrestle your data and information to the ground. You've got to get it organized and straightened up... and that's partly our interest in Benchling.
High-quality data is a prerequisite for advanced analytics, machine learning, and augmented intelligence. A first step is to ensure that data is centralized and standardized. An informatics platform must provide a data model that maps to your R&D workflows, and can be adjusted when necessary.
Scott Hamilton shared data visualization and analysis produced by connecting Benchling's Data Warehouse to SQL-based data visualization tools like Tableau and Power BI. Synlogic is using data visualization dashboards to track fermentation trends over time and across experiments, to track reference material over time, and even to monitor equipment usage by project.
We built pre-designed hot-swappable data visualizations and graphs that we could bring data in and out of really quickly.
5. Implement Benchling in one team - and others will follow
Implementing an R&D informatics platform can be daunting. Even switching from spreadsheet, email, and paper-based systems requires careful change management, user training, and enforcement. Benchling users were eager to share best practices for implementation, and ongoing maintenance of the R&D informatics platform.
Tiffany Chen outlined how Benchling was first implemented in a cell engineering team known as Rubius Prime. The team works all the way from DNA to engineered cell therapeutic, and is relied on by other biology groups at Rubius to develop prototype cell therapeutics. Rubius focused on achieving full adoption in this group before rolling out to other teams.
We started in a small team first... When they adopted it and found it very useful we expanded it to the other biology teams and they saw the usefulness of the entire system.
In the case of Synlogic, the Benchling Lab Notebook was rolled out to the entire company, but the Workflow Management module and Data Warehouse started out in Scott Hamilton's platform development group.
We got this program running, and we thought it was really great, and we thought we were done. Once other groups started to get wind of this, they wanted to, of course, get their data into the same systems.
In both cases, defining a clear functional need for Benchling in one group was helpful in demonstrating the usefulness of the platform to the whole organization. This was reinforced by internal power users, or “Benchling champions”, who devoted time to on-boarding new staff members and reinforcing good practices.
What comes next?
Benchling's User Forum brought together thought-leaders in the rapidly changing biotechnology industry. The presentations reinforced that life sciences R&D is becoming more complex and sophisticated, and is placing greater demands on the informatics infrastructure that supports it. As an industry, we're growing faster, and generating more data than ever - but extracting value from it poses a challenge. The benefits are clear for structuring, centralizing and automating data management. The difficulty is doing this in a way that's flexible enough to grow with the company, and easy to configure, integrate and extend.
At Benchling we believe we're uniquely placed to deal with these challenges. We're excited to work with this dynamic community, and to bring the group together through events like our User Forum.