During the SynBioBeta SF conference last October, Riffyn announced the launch of its cloud-based research process design software. The company’s mission to change the pace of research and development through software innovation has now reached new speeds through a key partnership with Novozymes, the largest provider of enzyme and microbial technologies worldwide. The industrial giant will deploy and develop Riffyn’s software, implementing it progressively across Novozymes’ 1000-person global R&D organization to support product development.
With Riffyn’s “design-first” approach to scientific experimentation and product development, Novozymes seeks to address the issue of their exponentially growing R&D data. The company needed real-time access to high-quality structured data sets in order to link, blend, analyze and visualize the information across technical teams. Riffyn’s proposal of a software as a service capable of eradicating data fragmentation and enabling access to integrative data analytics fit the brief perfectly.
“Novozymes is one of the most innovative industrial biotechnology companies in the world. We are excited to help them achieve unprecedented efficiency and effectiveness in their product development,” stated Timothy Gardner, Founder and CEO of Riffyn, in a recent press release. “The deep partnership with Novozymes is also a great boost to Riffyn’s own product development—helping us to rapidly expand our data acquisition and visualization capabilities.”
The gravity of data fragmentation and poor data management continues to increase as the speed of R&D data production accelerates and becomes more complex and abundant. It’s widely recognized by now that these problems lead to low experimental data quality, and puts up barriers for effective collaboration. By Riffyn’s analysis, these issues collectively cost global R&D organizations well over $100B per year in wasted research effort and delayed technology commercialization. This estimate is based on the year total R&D spend for process-based industries, considering a 25-75% error rate, described as R&D results that fail to reproduce, leading to wasted resources, failed tech transfers and missed discoveries.
Novozymes isn’t the only company that can and will leverage Riffyn’s solution. The company is already supporting groups as small as 2 scientists and as large as hundreds – moving rapidly to thousands thanks to Novozymes. Small labs and large enterprises have the same problems of data fragmentation, poor data quality, ambiguous process design, and barriers to collaboration. Riffyn was designed from the start to help small teams of scientists collaborate on experimental design and data sharing across physical and geographic boundaries.
Since Novozymes’ work encompasses areas as broad as agricultural yields, low-temperature washing, energy-efficient production, renewable fuel, among others, it’s no surprise that the company struggles with the massive amount of diverse data that they create. As stated by Søren Egestad, Director of Scientific Data Management and Analysis at Novozymes, in the company’s press release: “Riffyn delivers a vehicle to harvest data from our global lab processes in a structured yet flexible way. By means of Riffyn’s process oriented data model, we will be able to even further strengthen our lab flows. Riffyn helps us to secure a close connection between the lab process and the data collected and stored for analysis. In our perspective, Riffyn offers a new and differentiating paradigm for the lab side of our scientific data management activities.”
The key to tackling a company with processes as broad and complex as Novozymes’, according to the Riffyn team, was to recognize the similarities in how scientific R&D is performed, regardless of scientific specialty, according to Derek Gregg, the company’s Director of Business Development. “Riffyn’s experiment design and data handling capabilities take advantage of these similarities to collate and organize data onto repeatable input/output relationships across a process” says Gregg. “Riffyn acquires data through multiple modes — database queries, configurable file parsers, manual entry and an API — and transforms all this data, no matter the scientific discipline, into a common and repeatable input/output data pattern for standard statistical analysis and machine learning.”
So what’s next for Riffyn’s platform? Gregg states: “We’re very excited about the progress we’ve already made. Future development is focused on making our UI for data acquisition and experiment execution more automated, on new high-throughput screening tools, advanced statistical analytics and visualization, and upgrades to the versioning and change control capabilities for complex distributed projects.”