Digitalization Of The Lab: Rethinking The Lab In A Digital World

By Trish Meek, Senior Director, Portfolio Owner, and Jonathan Scott, Director of Product Management, Connected Science, Waters Corporation

At first glance, the laboratory of today looks remarkably like it did twenty years ago. While there is a higher level of automation particularly of routine and repetitive tasks, and there are certainly improvements to the instrument and software technology, the overall process often feels unchanged. However, a true revolution of today’s laboratory processes is on the horizon, advancing with increasing momentum with the help of groups such as BioPhorum. BioPhorum, is a collaboration across over 130 biopharmaceutical manufacturers and software and instrument vendors, which has published a manifesto envisioning the evolution of the QC (Quality Control) Lab of the Future, enabled by digital technology. Pharma 4.0 and digital transformation initiatives are transforming the manufacturing environment, but the laboratory has fallen behind. BioPhorum highlights the need to digitally enable laboratory processes and improve access and interoperability of scientific data. Laboratories rely on digital technology to support the scientific process, but these tools are often disconnected from one another. To truly harness artificial intelligence (AI) and benefit from tools like generative AI and machine learning, we must rethink how we create, curate, and access data. True digital transformation demands a reimagining of laboratory operations in the digital world.

In the past, organizations often implemented paperless lab projects. Going paperless, while a worthy goal both for its improved efficiency and environmental impact, often meant that organizations focused on digitization, also known as “paper-on-glass.” As Gartner describes, what digitization lacks is the opportunity to change processes themselves. Take the example of an electronic form that captures the same information as an existing paper form. The organization no longer needs to maintain the original paper record and information is immediately available through the digital record. While this example of digitalization enables the organization to eliminate the need to maintain paper records, the focus solely on being paperless has resulted in the greater opportunity being missed. Mere digitization does not optimize lab processes, improve scientific insight, or deliver new outcomes.

The Digitalization Journey: A New Era for Labs

The digitalization journey, on the other hand, heralds a new era for laboratories. Defined by Gartner as “the use of digital technologies to change a business model and provide new revenue and value-producing opportunities; it is the process of moving to a digital business.” According to an Accenture survey of 118 industry leaders, 72% of labs identified the ability to support new science and innovation as a key driver for digitalization. This shift requires a fundamental change in lab operations and better connectivity of scientific data across their organization. As companies digitally transform their end-to-end operations, it is critical that they focus on the entire process, not just on how to handle the data after it is created. For example, through smart instrument diagnostics, laboratories could detect an instrument or column failure before it occurs, significantly reducing downtime and improving productivity. Similarly, chromatographic peak integration, a challenge with complex matrices and large molecule workflows, could be revolutionized by AI. A recent paper by a team at Merck, Satwekar et al., proposed a deep learning algorithm for peak integration, potentially enabling true review by exception, freeing scientists to focus on science. Both examples highlight how we need to rethink our current processes to take advantage of digital technology.

BioPhorum’s Vision: Data at the Core

BioPhorum’s vision for the QC Lab of the Future includes three key business capabilities:

• supply management,
• automated testing
• test output management.

Central to these is data captured, integrated and searchable automatically. They go on to describe the data analytics infrastructure as a key “enabling dimension” of the QC Lab of the Future. Many organizations are adopting FAIR (Findable, Accessible, Interoperable, Reusable) data principles as the guiding principles of their data analytics infrastructure.

To solve their complex business challenges, organizations are developing enterprise data lakes and lake houses, aiming for comprehensive data accessibility. The reality is that it is difficult to create and manage a single enterprise data lake that enables insightful and timely decisions across all data challenges. The laboratory is a heterogeneous environment, and each instrument can have a different data format and integration mechanism. Building data pipelines to a central data lake takes time. Even after the infrastructure is complete, the data science team must collaborate with scientists to fully understand the problems they are addressing. This interaction can delay access to information because the questions scientists are trying to answer often depend on the specific instrument, sample analysis, and underlying biological processes. The complexity increases when the data is needed to ensure regulatory compliance and maintain product quality. The ideal solution would be a compliant ‘data pond’ for the laboratory environment, an “analytics-driven, purpose-built” solution specifically for laboratory data that enables real-time decision making. The data pond will provide numerous benefits over the enterprise data cloud. Often when data is transformed, it loses some of the contextual information about how the data was acquired, such as instrument conditions and process history, that can impact the validity of the data. Additionally, the data pond should allow visualization of the scientific data going beyond just calculated results. This data pond would also, of course, connect and pass data directly into the corporate data lakes, integrating curated lab data with manufacturing data from manufacturing execution systems (MES) and enterprise resource planning solutions (ERP).

Embracing the Future: A Digital Lab Evolution

There is no doubt that digital technology will fundamentally transform science. With the rapid pace of innovation in artificial intelligence, it is impossible to predict where this technology will go next. What is clear is that the laboratory is not optimized to take advantage of the increasing advancement in digital technology. Unless we evolve laboratory processes and the supporting data infrastructure, this will continue. To build the lab of the future, we must start by rethinking the lab in the digital world.