Sample and Workflow Management
A crucial precondition for extensive automation is to clearly identify the samples in all their different states. "Contextualizing the data related to the sample will be the next challenge. It is no longer sufficient to stick the barcode onto the tube once. Now it is a matter of tracking the states of the sample changing within the protocol". This could be accomplished in a few years, for example, with block chain, which is currently still too expensive. "In principle, however, this is the idea behind it: We continue to map the digital twin of the physical specimen. The next step towards a practically unmanned laboratory is the topic of workflows. Once I am able to configure my recipes or protocols step by step with all the dependencies on the devices and their respective states, all that is missing is a clever way to physically transport the samples. There are several possibilities, from linear guidance systems to robots, cobots or even drones.
Device Management
"Linking an electronic pipette to the cloud doesn't help much at first, but it does add real value once I can keep an eye on the condition of the pipette at all times. The fact that the researcher in charge and the laboratory manager can check at any time where the pipette is, in what condition it is and what its history is, adds interesting value in terms of topics such as calibration, quality management or audits", says Gill. "It becomes even more interesting if we are not only able to check the current actual state but can also turn the pipette into an extended arm of the software by loading it with the appropriate protocols. If we could load a protocol to be executed onto the pipette, this would almost turn the pipette into a small robot. In addition to the pipette's utilization, quality assurance in the sense of preventing errors will increasingly benefit from this. Documentation of the parameters is also an important point", Dr. Gill says: "Up to now, we have spent a lot of time manually documenting what we have done in the laboratory. We are now striving to ensure that no one in the laboratory has to worry about this anymore, because contextualized data is automatically recorded in the background. Thus, a relatively simple pipette will in future become an essential tool fulfilling a whole range of tasks from documentation to planning and execution to quality control."
Laboratory and Workflow Automation
At a time when more and more people are working from their home office, the question naturally arises as to whether this trend can also be transferred to laboratory work. Here, Gill says, there is a clear distinction to be made between various types of laboratories: "In research, extensive automation will always be difficult because there are no fixed protocols for repetitive tasks in place yet. This means that researchers work relatively much in the mode of trial & error and could at best reach a semi-automated level." To achieve total automation, highly complex tasks would have to be simplified to such an extent that sufficient standardization can be achieved. And the more one has to try out, the more difficult this becomes. For repetitive tasks, on the other hand, the degree of automation is basically dependent on economic considerations, says Gill: "When it comes to tests that are the same every time, such as the test for SARS-CoV-2, for which there is great demand, automation allows the cost per sample to be reduced."
A completely unmanned laboratory nevertheless remains an illusion, Gill is certain: "If, for example, the rotor of a centrifuge needs to be replaced, this cannot be fixed automatically. But with over-the-air updates and remote services, we are able to solve problems at the software level, play out optimizations or request maintenance tasks. Step by step, we will also increasingly integrate the instruments into workflows, for example to further improve documentation. The calibration process can also be automated. "Asset management not only can keep track of the history and current status, but also allows in the future a forecast of when the next calibration will be due. "Devices will soon be able to order spare parts before they break down, Gill is sure: "This is called predictive maintenance, comparable to the use-dependent warning messages on the display of modern cars. But for this to work reliably, devices and software must be perfectly aligned.” At Eppendorf, the focus of attention is therefore not only on the devices, but also on the Systems used in Laboratories. "Our strength is the digital portfolio with the corresponding application-related knowledge in the background. We are positioning ourselves in the orchestration of ecosystems and rely on an agnostic approach, so that we are able to integrate third-party systems and instruments into our software as well. This open approach to the systems is one of their strengths."