Automation in drug discovery will enable pharmaceutical companies to make better decisions faster. Automation is often believed to be new to pharma, but in actual fact, there has been a long history of automated systems in drug discovery. At first, automation in the laboratory was primarily used for drug development and production to aid high-throughput analysis. Since then, advances in technology have enabled the evolution of additional approaches to improve the precision, compliance and replicability of results.
Automation has revolutionised the laboratory environment as its ability to perform large volumes of repetitive tasks has helped to alleviate some modern-day pressures on scientists, and also increase reliability, throughput and reproducibility.
Automation promotes a less wasteful environment, by increasing efficiency in the laboratory, and speeds up drug discovery processes. However, the integration of automation results in new challenges that need to be considered. This article covers ways in which pharma are using automation and modern automated technology to accelerate drug discovery.
Table of Content
Why automate your lab?
Benefits of automating the lab
The stages of laboratory automation
Laboratory automation software and devices
Challenges of lab automation
The future of laboratory automation
Why automate your lab?
Opportunities to automate processes in the lab have grown as a result of improved technological capabilities. Nowadays, automated equipment is a key component of many laboratories and can be used across the entire workflow. Automation is no longer just about high throughput, but also about inter-connecting the laboratory environment to generate greater precision and increase the accuracy of results in R&D.
Benefits of lab automation:
Reduced costs
- Automation has enabled large-scale testing. It has also reduced utilities and equipment costs. Most notably, a large number of tests can be performed by a limited number of personnel, eliminating large amounts of manual labour.
Reduced human error
- Repetitive tasks that are usually manually done can be carried out with ease by automated solutions. Output data from machines can be streamed directly into a scientist’s electronic laboratory notebook. This minimises the risk of human errors that arise from manual recording. It also removes the repetitive aspects of experiments.
Unified devices
- Laboratory devices are usually isolated. Automation facilitates opportunities to connect lab processes. Remote controlling of devices can also enable experiments to be conducted anytime and anywhere.
Maximise data generated
- Not only does automation enable easier collection of data, it also can produce extra data that is not usually collected; for example, data pertaining to optimal experiment conditions. Forms of automation can allow companies to stream this data onto an online platform and use it to make important decisions about machine placement and experimental conditions.
Efficiency
- Eliminating the need for humans enables tasks to be performed faster. This in turn allows for quicker turnaround times for results.
Flexibility
- Laboratory automation systems can be customised to meet a given laboratory’s needs. Additionally, technologies can be added and removed as technology advances, allowing laboratories to evolve. Most importantly, it can save space in the lab.
Frees up time and resources
- Many laboratory tasks are time- consuming and tedious. Eliminating these tasks for scientists can save a lot of time and ensure their skills are used for other more specialised tasks. While automation has a big price tag, it can improve the overall effectiveness and efficiency of the laboratory which can translate into an increase in scientific progress and reduce the time to market.
- Many laboratory tasks are time- consuming and tedious. Eliminating these tasks for scientists can save a lot of time and ensure their skills are used for other more specialised tasks. While automation has a big price tag, it can improve the overall effectiveness and efficiency of the laboratory which can translate into an increase in scientific progress and reduce the time to market.
Stages of laboratory automation
Laboratory automation can be split into pre-analytical, analytical and post-analytical stages. It has been estimated that pre-analytical errors account for more than two-thirds of all laboratory errors. Mistakes in the analytical phase and post-analytical phase account for one-third of all laboratory errors.
Pre-analytical automation
The pre-analytical testing is the first phase in the laboratory workflow. It refers to any of the steps from sample/reagent storage to preparation and identification. Significant errors can occur during pre-analytical testing, including mislabelling, poor storage and incorrect volumes, which can impact further downstream analyses. Consequently, automating repetitive tasks during this stage can have a big impact on the accuracy and reliability of results. A combination of robotics, sensors and software are available to perform many pre-analytical tasks, from storing and labelling, to transporting and centrifuging. However, the more automated systems become, the harder it is to unravel errors.
Analytical automation
Although the pre-analytical and post-analytical phases present the greatest potential for quality improvement, analytical quality can still present major issues. This stage involves the actual laboratory testing or diagnostic processes. Early forms of lab automation were involved in the analytical stages of experimental processes. The first automated analyser – AutoAnalyzer – was produced in 1957 and used continuous flow analysis. It improved the number of samples that could be tested each hour. Common risks during this stage are the use of expired reagents, or the testing of samples on reagents that have not passed quality control (QC). Advanced diagnostic platforms can aid in ensuring compliance by providing automated software features to prevent the use of expired reagents and to track external QC. Of particular interest is process analytical technology (PAT). It is a powerful quality assurance tool, measuring QC in real-time. Over half of the top ten global pharmaceutical companies are now using or evaluating synTQ PAT-based production and development software from Optimal.
Post-analytical automation
The final part of the workflow involves data analysis and interpretation, sample management and reporting. Various automated equipment have been designed to archive samples after testing. This reduces errors, such as mislabelling or incorrect storage. Samples can then be readily retrieved if further analysis is required. The main errors during this phase are those that occur during data entry and the manual transcription of results. Implementing a laboratory information management system (LIMS) can eliminate data entry transcription errors by automatically releasing results directly to the LIMS.
Laboratory automation software and devices
Different labs comprise of many different machines, software and devices that vary depending on laboratory preference. Additionally, different research specifications have particular preferences for software solutions. Here, we describe just some of the devices and software utilised by pharma that are transforming laboratory automation.
Automated liquid handling systems
Liquid handling plays a pivotal role within laboratories. Manually, this process can be time-consuming and impractical. Consequently, there is a strong demand for automated liquid handling robots. The simplest machines dispense a specific volume of liquid from a motorised pipette, whereas more complex systems can manipulate the position of the dispenser and/or integrate additional laboratory devices, such as centrifuges and microplate readers. These systems offer precise sample preparation for high throughput screening and bioassays. For example, up to ten thousand samples per day can be performed to determine the activity against a biological target of interest. Digital microfluidics is a promising liquid handling technology that has been shown to automate biological experimentations in a low-cost, rapid and programmable manner with applications for drug discovery.
Top automated liquid handling robot companies:
- Agilent Technologies
- Aurora Biomed
- Beckman Coulter Life Sciences
- Biotage
- BrandTech Scientific
- CEM Corporation
- Eppendorf
- Hamilton Company
- Hudson Robotics
- INTEGRA Biosciences
- ProGroup Instrument Corporation
- Shimadzu
- Sirius Automation Group
- Thermo Fisher Scientific
- Gilson
- MDZ Automation
Source: labcompare.com
Electronic Lab Notebook (ELN)
The most widely known form of digitisation in the scientific community is the electronic lab notebook or ELN. In its simplest form, an ELN can be considered an electronic embodiment of what is currently being done in a paper laboratory notebook. It facilitates the workflow and helps researchers document experiments. ELNs tend to be more personalised and flexible as they can be changed by individual researchers. They are generally better suited for research and discovery environments. Selecting an ELN is dependent on many factors, including consideration of software integration, legal issues, costs and ability to run on other devices e.g. mobile or tablet. Several pharma companies have implemented ELNs, including Johnson & Johnson, Merck, GSK and AstraZeneca. Many have taken a hybrid approach to the new technology – using ELNS for data entry followed by printouts.
Top Electronic Lab Notebook (ELN) companies:
- PerkinElmer
- Dassault Systemes
- Waters
- IDBS
- Abbott
- Agilent Technologies
- Thermo Fisher Scientific
- Bio-ITech BV
- KineMatik
- LabWare
Source: Meticulous Market Research. Electronic Lab Notebook (ELN) Market – Global Opportunity Analysis and Industry Forecast (2019-2025)
Laboratory Information Management System (LIMS)
Laboratory information management systems (LIMS) play a key role in library automation. LIMS is a software that allows the effective management of samples and associated data. Using a LIMS can automate workflows, integrate instruments and manage associated information. An efficient LIMS is required to ensure that the quality of data management of laboratory results meets that of the automated laboratory processes. Choosing a LIMS involves selecting the appropriate database engine, ease of use, quality of technical support and training programs. The utilisation of existing hardware or software also needs to be considered. Other factors include the flexibility of changes to the system, the user interface, vendor expertise and, of course, budget. LIMS were first introduced into pharma over 30 years ago. Since then, they have evolved and become pivotal within laboratory evolution. There are three alternatives: a custom-built system, a customised generic system or a commercial off-the-shelf solutions (COTS). Many of these systems undergo extensive customisation to satisfy pharma business requirements. The LIMS market is forecast to be the fastest-growing market from 2020 to 2025.
Top Laboratory Information Management System (LIMS) companies:
- Thermo Scientific
- LabWare
- CrelioHealth
- Benchling
- Labguru
- CompuGroup Medical
- Abbott Informatics
- LabVantage
- Sunquest Laboratory
- SoftLab
Source: g2.com
The challenges of lab automation
Many manual tasks have now been partially or completely replaced by automated instrumentation. Nonetheless, automation is not as simple as replacing humans with machines; in fact, it presents many challenges. Here are just some of the challenges experienced by pharma companies when considering and implementing automation.
Skills gap and loss
Scientists monitoring the software need to be trained, and as technology advances, the need for further training increases. The task of monitoring software may be different from traditional manual tasks so retraining on how to use new software can present a major challenge to automation. Additionally, as new technology is adopted, old technology is discarded. This results in the loss of skills and knowledge surrounding the old system. Previous technical knowledge is often useful for troubleshooting problems with new equipment. Therefore, it is important to make an effort to maintain these skills.
Learning curve
The learning curve is a period of time that is required for lab technicians to become accustomed with novel technology before it can become fully optimised. Automation changes how people interact with testing. Lab technicians are trained in handling specimens rather than software. There are different skill sets involved in managing software. Judgement calls must be made about releasing test results, monitoring equipment, performing quality control testing and deciding whether to perform maintenance. Figuring out these new interactions can be challenging, particularly as workflows may change.
High short term costs
The investment for automation is inevitably associated with an initial escalation of costs for accommodating the project, system installation and new hardware. This may be an issue for some facilities. Therefore, a negotiation may be necessary to clearly illustrate the possible return on investment achievable by shifting towards automation. This must be accompanied with a reliable financial plan for expenses and projections of revenue.
Infrastructure constraints
Space requirements and infrastructure constraints are major issues for implementing automation. Accommodating equipment in a pre-existing environment, particularly when the laboratory is not built for purpose, can be a challenge. Most accommodation must install these new systems whilst maintaining their service obligations. Implementing automation in a new open space is ideal but is rarely an option.
The future of laboratory automation
The use of automation can alleviate the burden of many time-intensive tasks. While the widespread application of these advanced technologies can create challenges for laboratories, there are many opportunities to be gained from implementing such systems.
In recent years, intelligent process automation (IPA) has become the latest buzzword within the automation space. IPA is the latest wave in automation technologies, whereby automatic technologies converge to produce dramatically improved capabilities. This convergence includes smart workflow and machine learning. Another growing trend is ethics relating to automation. The impact of automation will have both a professional and social impact, and play a more prominent role in decisions. It is expected that additional legislation towards newer autonomous tools will come to fruition, leading to a wider acceptance of automation technology.
Finally, a call to action – attitudes towards automation and other advanced technologies must change. We must move away from the assumption that all humans will be replaced by machines. Instead, we should view them as collaborators that support of human work. If machines carry out all of the tedious and repetitive work, humans can focus on the higher cognitive tasks. Integrating automation will create new jobs, and in return, fundamentally change other roles. We must think about how we can work synergistically alongside these technologies, rather than considering them as a threat to our existence. These collaborative efforts will change the lab automation industry for the better.
