Google Cloud + TetraScience: Partnering to catalyze scientific AI innovation
Learn more
We welcome the opportunity to speak with Andrew Skalkin, CEO and Founder of Datagrok, a powerful ML/AI and Data Visualization provider, who is also a member of the Tetra Partner Network.
Datagrok delivers major improvements in performance, scalability, ease of deployment, integration, and portability over competing solutions. Out-of-the-box, we offer rich exploratory data analysis capabilities, machine learning and predictive modeling, data augmentation, collaboration, knowledge dissemination, and the use of AI to learn from data and user actions.
In addition, the open architecture and component-based design of the platform allow new functionality to be rapidly developed and deployed as plugins. These open-source extensions include first-class support for cheminformatics and bioinformatics, allowing our customers to analyze millions of molecules interactively in the browser, as well as build their own applications.
Thanks to this powerful technology, Datagrok is a partner of choice for several leading pharma companies. Working closely with them, we have developed applications spanning multiple domains - from early drug design and discovery for small molecules and biologicals all the way to analyzing clinical and RWE data.
We are thrilled to have found a partner in TetraScience to help our customers derive faster and better insights from their experimental data. The fact that TetraScience has an open R&D data cloud aligns well with our open source extensions and mentality.
It’s all about data, a critical element in today’s world. The speed at which insight can be derived, and the time it takes to act on that insight is paramount. However, many obstacles prevent biopharma companies from unlocking the full potential of their data. The very organizational structure of such companies often creates barriers for data and initiatives, hampering collaboration efforts. The data lives in multiple, often isolated, silos. It is highly complex and multi-dimensional, creating major challenges in storage, integration, analysis, visualization, and reporting. It also spans multiple domains and requires domain-specific terminology, intelligence, and models.
Datagrok was purpose-built to address these challenges by blending biochemical intelligence with high-performance computing and leveraging technological advancements in modern browsers. The result is a fast and intuitive platform for visual data exploration of life science data. What also sets Datagrok apart is its flexibility. Through its many plugins and extensions, users can connect to virtually any data source, visualize data on the fly, compute various properties, build or apply predictive models using state-of-the-art AI techniques; or they can build their own fit-for-purpose applications on top of the platform.
When Alan Millar, Vice President of Business Development for the Tetra Partner Network, and I first spoke, we immediately recognized the tremendous potential that this partnership unlocks for our customers. While Datagrok can connect to any data source, we still need engineered data to get to scientific insights.Through Tetra Data, Datagrok can access data in a centralized, clean, and harmonized way. The combination is compelling as it transforms how experimental data goes from its raw form to actionable insights, improves modeling and workflows, and ultimately helps accelerate the development of therapeutics.
Consider a high-throughput screening campaign to identify new drug candidates. The interesting thing about HTS is that it typically involves a variety of lab equipment, and data resides in multiple internal systems. For scientists to fully exploit their HTS data, they need to be able to interactively visualize and analyze the data, test hypotheses, get insights, and act on them faster and better than before. With Datagrok, they are able to achieve all of the above: navigate through sizeable plate collections with incredible speed and ease; cross-link and correlate HTS data to instantly visualize hits, leads, and responses; run scientific computations or predict outcomes; visualize the results with interactive viewers, and share these results with anyone in an organization.
While Datagrok’s HTS browser is a generic solution, integration with customer databases takes a lot of effort due to the differences in data representation. Being able to build applications against engineered data through easily accessible API’s is a game-changer.
This is exactly what Tetra brings to the table. It provides a universally adoptable, comprehensive format, containing raw data, processed results, and context. With Tetra, we can build just one application against Tetra Data - and all Tetra customers are now able to use the app without a significant amount of rework. All they need is to mark their company data with the appropriate metadata!
But there’s more. The application would become a part of Datagrok’s rich open-source ecosystem, offering our clients lower cost of ownership, the ability to customize the solution to their needs, and access to new features developed by the global community. The possibilities for pre-competitive collaborations are endless, and we are seeing a lot of interest from our customers already.