Recent estimates indicate that the value of the global biotechnology market will exceed USD 775.2 billion by 2024, growing at an impressive CAGR of 9.9%. Biotechnology covers a variety of verticals, including bioagriculture, bioservices, bioindustrial and biopharmaceutical. The last of these makes up the lion’s share of the biotechnology market and is an area of growing investment.

Biopharmaceutical companies continue to be on the lookout for technological improvements to increase their efficiency and bring products to market quicker. Automation, and the integration of other technological systems via the Internet of Things (IoT) and analytics platforms like machine learning, will form a major part of the future of scientific laboratory work.

A particular area of focus is bioprocess development, where the conditions for the optimum production of a new drug product are developed. This is a time-consuming and complicated process that can be dramatically improved using modern software and informatics solutions. High throughput technologies such as microbioreactor systems are already recognized as valuable additions to development labs but there are still challenges to be overcome before these ‘islands of automation’ can be used to their full potential.

Why is automation needed?

Biotechnology labs in general, and biopharmaceutical development labs in particular, are often a study in contrasts between highly sophisticated scientific methods and instruments on the one hand and paper lab notebooks, binders and repetitive manual tasks on the other. In addition to often quoted efficiency improvement gains, one of the other key objectives of automation is to improve the reproducibility of experimental results. A Nature survey from 2016 revealed that, of the 1,576 researchers who responded, more than 70 percent had tried and failed to reproduce someone else’s experiment. Even more surprisingly, more than 50 percent had failed to reproduce their own experiments. While it isn’t possible to attribute the problems with reproducibility to a single factor, it’s easy to see how eliminating sources of human error through automation can help.

The rise of the lab robot

The first thing many people think of when it comes to lab automation is robots. The use of robots in the lab isn’t a new concept; commercial robotic systems first emerged in the 1980s. The initial purpose of these systems was to replace manual tasks such as liquid handling/pipetting and plate management. Further technological advances meant that robots could connect to corporate inventories and sample management and dispensing systems, meaning that scientists could design experiments and then effectively walk away and let the automation do the work.

Many of the new instruments coming to market are IoT enabled, and we are seeing a rise in vendors adopting these concepts to help deliver better services, such as automated preventative maintenance and reagents delivered just in time (JIT). By making it possible to target the right information at the right audience, IoT enabled instruments are not only easier to use, but can also help ensure data integrity and give early warning indicators of potential problems.

Many of the robotic systems designed for lab use were at first difficult to operate and often prohibitively expensive for all but the most generous research budgets. This is changing with the introduction of more affordable technology such as Opentrons, and vendors are working to make robotic systems more user friendly. Programming a robot used to require specialist knowledge and was typically the responsibility of one or more computer-savvy scientists or IT support staff. Now, almost anyone in the lab can create robust and reproducible workflows for automated systems to execute.

What comes next?

Once a fully connected and automated environment has been established, the focus can then shift to the next phase of development—a self-monitoring and regulating closed system that can make decisions on what task should be undertaken next based on the current status of an experiment. As analytical instruments become ever more sophisticated, and the wealth of data produced at each experimental stage increases, informatics solutions need to keep pace so that the data can be turned into useful information.

Monitoring analytical instruments in real time, gathering data and performing automated analysis and feedback loops all must be considered at this stage, and models can even be developed and trained to allow systems to control themselves based on real-time patterns that they are seeing. Companies such as Lab Genius and Ginkgo Bioworks are already applying machine learning algorithms to protein engineering and synthetic biology with great potential for both therapeutic and consumer product development.

The foundations of this technical revolution are IoT and robotics—but their routine use in laboratories and manufacturing contexts will require considerable optimization and rigorous testing. Biological systems are notoriously complex and unpredictable and processes that work well at small scale can be difficult, if not impossible, to scale up to production level, as companies like Amyris have discovered to their cost.

The final part of the system to consider for automation is data analysis and reporting, both of which are critical parts of the decision-making process for scientists. Typically, the process for data aggregation, analysis and reporting—despite the automation of the instruments—is still manual for many organizations. Great strides are being made to accelerate analysis and reporting automation, with product concepts of alerting, automated decision trees based on business rules and machine learning, and research and development ‘project progress’ reporting all starting to emerge.

For analysis aspects, it’s important to make sure that all the different system parts are integrated at the data level, which can then allow the automation and analysis to occur.

The need to both increase efficiency and reduce time to bring new products to market is forcing companies to reconsider the ways they interact with and leverage technology. By considering automation in this holistic manner we can see great opportunities for new technology and a streamlining of research and development. The possibilities of making new products cheaper, faster, and with higher quality than ever before is an exciting prospect to look forward to.