New systems with new capabilities promise to upgrade R&D lab operations, but they won’t solve the needs of all researchers.
Lab automation systems are used in R&D laboratories for a wide range of operational applications and purposes. According to a recent reader survey performed by the editors of R&D Magazine in late-2014, the top three applications include to improve the accuracy of lab operations (selected by 61% of the readers), to improve lab productivity (58%) and to ensure the reliability of the lab operations (48%).
Lab automation is used by significantly more life science researchers than physical science researchers and, as a result, there are higher response rates to our survey by life scientists than physical scientists. For example, about 73% of the life scientists who responded to the survey considered accuracy their largest application, while about 54% of the physical scientists considered accuracy their largest application (for a net average of 61%).
Lab automation generally eliminates the human inconsistencies introduced into lab operations, especially those operations that are heavily repetitive, time consuming and have large sample sizes. As lab technicians continue to do the same operations for long periods of time, they become fatigued, and the consistency of their manual operations drops. Lab automation systems also can work 24/7 and, lately, have been designed and configured to work continuously over weekends when sample holders are loaded on a Friday afternoon. Researchers can then come into their lab on a Monday morning and download the results of the lab automation procedures completed over the weekend.
An interesting statistic arises when considering sample processing accuracy when revealed by both researchers and equipment suppliers. While the data noted above reveals researchers’ importance of the accuracy of lab automation systems (61%), lab automation equipment suppliers take a more aggressive approach (chosen by about 75%). Equipment suppliers also indicate reliability, repeatability and operational difficulty are among the most important issues in considering lab automation systems (with similar ratings to accuracy); while researchers consider these issues important, but not by the same level of concern.
Defining lab automation
Lab automation is a broad term encompassing a wide range of equipment, products, software and procedures. The largest number of lab automation systems used by researchers (according to our reader survey) include centrifuges (selected by 35% of the respondents), stirrers/shakers/mixers (33%), imaging systems (32%), weighing systems (32%) and application-specific systems (31%). True to the specific discipline bias noted above, life scientists indicated they had and used significantly more (by approximately 20%) centrifuges and stirrers/shakers/mixers in their labs than physical scientists. But, noting the specific lab operations likely to be employed by each of these general disciplines, physical scientists indicated they had and used more (by approximately 5%) application-specific systems and imaging systems than the life scientists.
Application-specific lab automation systems are custom designed and configured for specific applications. Generally, commercial off-the-shelf (OTS) systems don’t exist for many often complicated and possibly proprietary lab operations; and to eliminate the inconsistencies and shorten the time researchers and technicians might have to employ to get their desired test results, these application-specific systems are designed and constructed. Due to the larger potential market size, there’s a wider range of OTS lab automation systems available for life scientists than for physical scientists.
It should also be noted that for some general lab equipment systems, like centrifuges and stirrers/shakers/mixers, the larger versions are considered by many researchers as lab automation systems. But, the smaller-capacity versions may not be considered (or used) as lab automation systems by other researchers. These OTS products are considered as lab automation systems when they have the ability to simultaneously process multiple samples and large volumes. Some floor-mounted, free-standing centrifuges, for example, can simultaneously process up to 6 L or more of blood samples.
Many of these systems may also have dedicated commercial or application-specific lab automation accessories integrated into their operating procedures. A large number of life scientists (41%), for example, also employ autosampling systems as part of their integrated lab automation environment. Many fewer physical scientists (9%) employ autosamplers in their lab automation procedures.
The most widely known and acknowledged applications of current lab automation design and development efforts are in lab robotics. These systems are used in high-throughput screening, combinatorial chemistry, clinical and analytical testing, medical diagnostics, large-scale biorepositories, chemical and materials processing systems and storage systems, along with many other applications.
A concept referred to as total lab automation (TLA) is now being developed in a small number of specialty labs (mostly high-volume clinical or biomedical applications) that consists of specimen sorters, automation systems, robotic transporters and analytical analyzers. These systems can have a large impact on automating the testing and further increasing the lab’s productivity by reducing most manual operations.
Life sciences
A number of companies offer automated liquid handling workstations for life science applications and workflows. One of these is Agilent Technologies’ Encore Multispan Liquid Handling System. “This system allows simultaneous pipetting from multiple positions on the workstation’s deck, which is ideal for hit picking or plate reformatting,” says Kasia Proctor, a senior product manager at Agilent, Santa Clara, Calif. “The Encore also has software improvements that enable its integration of more lab automation devices,” she says. “While lab automation functionalities can be quite complicated, fewer organizations are keeping lab automation engineers on staff. The expectation now is for scientists to be able to set up their own automation systems, making it yet another piece of equipment they need to learn. So, the easier it is to use the integrated software, the more approachable the entire system becomes. We offer ‘Forms’ that are basically a user interface for our protocols. We have a number of these ‘Forms’ that are pre-designed and ready to go with the basic system, or they can also be custom designed to make the final user interface much simpler.”
|
Another integrated lab automation workstation is PerkinElmer’s Janus BioTX, which enables consistent small-scale purification and sample preparation operations for analytical protein characterization that’s required to support quality-by-design experimentation in both upstream and downstream processes. “While these processes are important parts of process development, the design of experiment procedures can be time consuming, and sample analyses can easily exceed the capacity of most labs,” says Brian Kim, president of PerkinElmer’s Life Sciences & Technology Div., Hopkinton, Mass. “With the BioTX, we can accelerate this process to help life science researchers make critical decisions earlier in their development work flows.” This workstation enables column, tip and batch chromatography modes, which also eliminates the need for multiple instruments.
The BioTX comes in two versions, the Pro and Pro Plus workstations. Pre-programmed methods automate a variety of commercially available ion exchange and affinity chromatography solutions, including miniaturized chromatography columns, resin-packed pipette tips and resin-filled microtiter vacuum filtration plates. The BioTX Pro Plus presents a platform with 10X advantages in throughput and 5X reduction in protein mass requirements over higher-scale fast protein liquid chromatography platforms. The capability of generating predictive data for a variety of process development experiments with the BioTX presents a very usable platform that can help accelerate the development of biotherapeutic proteins.
Important features
Different system characteristics have different relative importances to researchers considering the purchase of new lab automation systems. According to the R&D Magazine reader survey, the most important feature in this lab automation system purchasing process is reliability. Researcher respondents rated reliability as 1.39, when based on a scale of 1 (very important) and 5 (unimportant) (see chart on pg 9). The values for life science and physical science responses was slight, but still noticeably different, with 1.35 and 1.56 values, respectively. Ease-of-use had similar ratings with a 1.48 overall rating, and life and physical science response ratings of 1.28 and 1.58. The least important characteristic was in the specific brand selection of lab automation equipment, with an overall value rating of 2.75, and life and physical science values of 2.75 and 2.66.
The other characteristics, in increasing order of preferences, include maximum capacity (2.22), networking (2.12), integration capabilities (1.91), initial cost (1.86), software (1.79), service and support (1.68) and flexibility (1.66). In all these, except for flexibility, the life scientists indicated the specific characteristic value noted revealed a higher importance than that chosen by the physical scientists.
Limitations
Researchers, both in life and physical science work environments, can’t automate all the operations in their labs for a variety of reasons. These reasons include sophisticated technologies and relatively mundane issues. The most noted limitations in the survey related to sample labeling systems, DNA sequencing, sample preparation and fluid handing areas.
“We wanted to use the ‘fancy’ OTS labeling systems, but were disappointed with the one we bought that would no longer print labels on good labeling material because it was past its expiration date,” says a life science researcher in Singapore. “We had to keep labeling the old way with tape and Sharpies, which didn’t work well in our -80 C freezers.” Another academic researcher at the Univ. of Southern California revealed that none of the available OTS labeling systems were accurate enough for his work.
Another academic at the Univ. of Punjab, India, noted he would like to automate the preparation of growth medium for their experiments, but is limited in the lack of appropriate equipment availability. “We want to automate the preparation of growth medium so that we can minimize the chances of contamination in our tissue culture experiments,” he says.
Other limitations occur because of inherent system issues. “Some of our older equipment can’t be automated due to the incompatible older operating system interface connections,” says a manager of water quality systems.
“We measure the volume of irregular objects, but no fluid- or physical-based automation system has sufficient accuracy and repeatability,” says a consulting engineer.
|
Costs are always a limitation as well, even when the desired specific lab automation systems are available. “The cost for a system autosampler is too high for the sample volume,” says a public health chemist. “We can’t automate our HPLC systems due to the high costs,” says a Univ. of Georgia lab manager. And “we’d like to automate the calibration process for our vacuum gauges, but currently can’t due to budget issues and staffing,” says a product manager in Utah. “We can’t automate our welding shop due to the high cost of the welding equipment and the instability of our electrical power,” says a department head at the Univ. of Eastern Philippines.
Ultimately, sample preparation or process complexity is also a deciding issue that limits the implementation of a lab automation process. “We prepare solid catfish samples for flavor analyses,” says a physiologist at the USDA. “The catfish must be purged and heated, and the flavors and steam condensed and transferred into a vial for automated SPME analyses—automation systems for these operations don’t exist.”
“In the plating of live cells, this equipment is hard to keep clean and sterile on an automated platform without a large capital investment,” says a graduate student at Northwestern Univ. “We work with a variety of polymerization processes, some are batch and some are semi-continuous, and we’d also like time-line data recordings of these, as well—OTS lab automation systems for these operations don’t exist,” says a polymer scientist in Michigan.
Software in the loop
Lab automation obviously takes many forms with information on increases in productivity often coming from the data captured by the analytical systems themselves. Behind the scenes of the hardware operations are automated data capture and reporting systems that help ward off system failures, while also scheduling things like routine maintenance. “Smart instrument control software, such as Waters ACQUITY Console Software, can self-diagnose performance problems, log errors, count detector lamp hours, count liters of solvent pumped and inject cycles and more,” says Elizabeth Hodgdon, senior product marketing manager, Waters Corp., Milford, Mass. “When a system failure occurs, the software can present this information in graphical forms so service support specialists can troubleshoot the equipment and bring it back up to speed faster and prevent further downtime. More specifically, the software generates status files that give operators running snapshots of the system and its overall usage state.”
Technology’s future
Lab automation systems are clearly in demand. More than half of the researchers queried in our survey indicated they’ve increased the number/level of lab automation systems in their labs over the past three years. Of those stating they’ve increased their lab automation systems, approximately half noted they’ve increased their automation systems by more than 15%. Similarly, more than half of the survey respondents revealed they plan to increase the number/level of lab automation systems in their labs in the future. Of those researchers expecting to increase their lab automation systems, more than half stated they would make those increases within the next 12 months.
Other than cost, there are relatively few issues with existing lab automation systems. As noted above, existing lab automation systems don’t satisfy all the researchers’ lab automation needs; but of those automation systems that already exist, there are relatively few concerns or issues. The biggest issue with current lab automation systems is their initial cost, which was indicated by 47% of the survey respondents (57% life science and 40% physical science). The next closest issue was ease-of-use with a 33% response rate. These two concerns dominated all other survey-listed issues, with less than 20% response rates each for accuracy, contamination, liquid handling, operating costs, power requirements, reliability, safety and size issues.