August 2, 2022
by TetraScience,
ACD/Labs is partnering with TetraScience, the Scientific Data Cloud company, to help pharmaceutical companies increase scientific data effectiveness. We invited Simon Meffan-Main, Ph.D., and Alan Millar, Ph.D., Vice Presidents, Tetra Partner Network, TetraScience, to share some insights on cloud computing in the pharmaceutical industry.
Twenty years ago, pharmaceutical IT departments started connecting individual laboratory instruments to a server so that they could aggregate their scientific data and gain better insights into scientific workflows. However, each connection had to be custom configured or was part of a single client-server system from one manufacturer. The systems also needed to be updated when the instrument software was updated. This produced enormous technical debt and ongoing maintenance for pharmaceutical customer IT departments. Additionally, this model proved neither scalable nor agile enough to easily incorporate new testing modalities, elastically scale to incorporate new visualization tools to make sense of data, or rapidly reconfigure workflows.
With these challenges, pharmaceutical companies have begun moving their laboratory solutions to the cloud to enable scale, automation, and increasingly, data science. Initially some of the solutions were home-built using services from cloud vendors such as Amazon, Microsoft, and Google. Increasingly, there are vendor-produced cloud lab informatics tools and applications. Do these tools fulfill the promise of lab of the future or do they risk creating another set of silos?
In this article, we share what’s possible with cloud technology for pharmaceutical customers and what’s at stake if the wrong strategy is chosen.
Cloud Possible: Engineered Scientific Data vs. Non-Engineered
Today, only a small fraction of laboratory instruments, informatics tools, and applications connect in a way that makes scientific data liquid and available to scientists who need it. Within a scientific workflow, each system connects to its own tech stack, with unique data models, application workflows, and storage locations. The data moves, but that’s not enough. Gaining insights beyond the ‘batch’ is difficult, if not impossible.
The current cloud revolution in pharma has the potential to solve this problem. With a well designed cloud tech stack the data can be instantly made available where it’s needed. However, that too, is not enough without engineering the scientific data so that it can be consumed by any other applications and tools in the ecosystem. This is where the real magic begins! When scientific data is engineered not just for one scientist but for many, it becomes reusable for any data consumer interfacing with the cloud platform. This reuse and actionability of data is unleashing a new era of innovation in pharma that is accelerating the pace of scientific discovery.
Furthermore, with the variety of scientific endeavors necessary for innovation in R&D it is not possible for one technology provider to do-it-all. This is where integration of informatics technologies and partnership between vendors becomes important to best serve the scientific community. The partnership between TetraScience and ACD/Labs is one such example. ACD/Labs’ unequalled capabilities to handle analytical data in all major vendor formats for various techniques—from processing and interpretation to storage and chemically intelligent search and visualization—is a perfect compliment to Tetrascience’s ability to harmonize, move, and make scientific data searchable in its cloud platform.
TetraScience has focused all our efforts on engineering scientific data, …so that customers can unify the data silos and use any of their scientific data beyond the initial experiment.
TetraScience has focused all our efforts on engineering scientific data, which we call Tetra Data, so that customers can unify the data silos and use any of their scientific data beyond the initial experiment.
- Tetra Data engineers both raw and primary scientific data from any of the hundreds of unique data formats produced by laboratory instruments and applications into data models that scientists can search, share, collaborate with, and easily move between applications and workflows. This is what we call data liquidity—the ability to connect an entire laboratory and company ecosystem. However, in order to manage and search this data intelligently, the raw data results need to be interpreted and processed, and this is where the ACD/Labs partnership is crucial to the success of this digital transformation.
- Tetra Data is compliant and FAIR (Findable, Accessible, Interoperable, Reusable) and can be used immediately by scientists and data scientists for data visualization and advanced analytics such as AI/ML.
- Legacy data can be engineered into Tetra Data and can be used, for example, to help identify new drug targets or fed into learning models.
Without a cloud strategy that includes scientific data engineering and the capabilities of ACD/Labs, pharmaceutical customers will still be unable to utilize the full value of their data. Investing in multiple cloud-first platforms without data engineering creates new types of silos because the data remains trapped in that platform and is not reusable. If scientific data is engineered, then no matter what analytical technologies or lab informatics platforms a customer chooses, the result will be persistent, future proof, and accelerative for scientific outcomes and insights.
Cloud Possible: Open System vs. Closed System
The pharmaceutical industry is one of the last industries to replatform its data to the cloud. With good reason! Regulatory requirements and the sheer volume and complexity of scientific data produced by thousands of vendors are staggering. Moving scientific data to the cloud is a journey best made with deep knowledge about whether the cloud platform is open or closed.
- An open cloud platform, or ecosystem, creates true network effects where data, from any source, created on any node in the network, can be reused by any other node in a one-to-many, or many-to-many relationship.
- In an open system, other vendors are encouraged to partner, to accelerate common customer outcomes, with each contributing their unique expertise to problem solving. This can include, for example, automating workflows so that customers can gain scientific insights and operational efficiencies.
- An open system supports scientific innovation because it’s designed for customers whose research needs change. For pharma customers using an open platform, it’s easy to accommodate new modalities, lab informatics, applications, or instruments.
In a closed system, pharmaceutical customers will still be investing in proprietary technologies that inhibit the sharing and potential value of their scientific data. With an open cloud platform every supplier in the customer’s ecosystem can collaborate to solve customer challenges. The Tetra Partner Network is vendor agnostic, which allows us to partner with any supplier while retaining an unrelenting focus on scientific data, which belongs to the customer!
Cloud Possible: Purpose Built vs. One-Size-Fits-All
General purpose cloud data platforms and tools are not built or optimized for scientific data or to handle the diversity of sources (instruments, software, collaborators) in structured and unstructured formats. They are then also not equipped to facilitate the complexity of pharmaceutical use cases, workflows, and requirements of doing science at scale. In addition to scientific inquiry, the key to data science is not the quantity of data but quality of data being injected into the platform. And as pharma has the largest number of instruments and software systems of any field of inquiry, the modalities of data are extremely diverse and complex.
…our technologies have been built for exactly the type of use cases scientists need every day in their labs.
This is where TetraScience and the ACD/Labs partnership comes in and truly helps customers, because our technologies have been built for exactly the type of use cases scientists need every day in their labs. Examples of how a purpose-built platform supports pharmaceutical customers include:
- A single small-molecule workflow can comprise dozens of sequential phases, each with many iterative steps that consume and produce data. As workflows proceed, they fork, reduplicate, and may transition among multiple organizations—different researchers, instruments, and protocols.
- Pharma customers often distribute research across geographies or in collaboration with external partners such as CROs and CDMOs.
- In development, QA/QC, and manufacturing, new data sources, formats, and workflows are added. Customers require oversight, and validation and compliance with regulatory requirements.
While general-purpose platforms can move data to the cloud and provide access to data as text and images, they are not designed to integrate and harmonize diverse scientific data formats to support customer workflows and requirements; or provide access to analytical data in a manner that is intuitive for scientists to review and re-use. With platforms designed specifically for chemical and analytical data, every investment made to move that data to the cloud and provide searchable access will continue to support scientific research, analysis, and outcomes.
Conclusion
There are no shortcuts on the path to creating enduring value for scientific data. By partnering with companies like ACD/Labs and TetraScience who are committed to creating unrestricted innovation and outcomes in the pharmaceutical industry, we can create a new era in laboratory science.
For more information, read the press release announcing our partnership here.
Guest Authors
Simon Meffan-Main
Vice President, TetraScience
Alan Millar
Vice President, TetraScience