Updated: Aug 5, 2020
Why scientists should embrace laboratory automation like our lives depend on it
If you are a scientist in the pharmaceutical industry, chances are that you have come across some sort of laboratory automation. There is a smaller chance that you have used this technology yourself, and maybe the lab you work for has not yet jumped on this bandwagon. However it’s certain that lab automation will only be getting more prolific in this industry, so the probability that these core lab automation technologies will not find their way into your work experience is practically zero.
Lab automation encompasses a wide variety of devices and techniques to provide consistency in the way that an experiment process is carried out. Introducing these sophisticated engineering controls into the laboratory helps gain confidence in the research assets and data produced. This precision and reliability is why automation has exploded in the manufacturing industry. You also see this technology being championed in industry where precision and repeatability is crucial, like in surgery. Since pharmaceutical research shares these same needs for precision and attention to detail, the use of automation can lead to better science.
Automation in labs most commonly takes the form of stand-alone robotic devices used by scientist to automate portions of the workflow, and some devices can even automate an entire process. Taking this concept even further, many labs are pushing the limits to build larger integrated systems in which several different devices are connected into a fully automated workflow. Freedom from the manual labor that these automated devices can handle gives more time to plan experiments. This means that scientists are spending less time doing repetitive labor, and they can get more experiments done as a result.
So automation is meant to mimic what a human scientist would do, why is it so much better? Not only do these instruments have a track record of outperforming human benchwork with lower error rate, but they are inherently more traceable and accountable. Performance can be readily measured, calibrated, and documented. This is an important concept to consider when we are talking about discovering medication. Machine logs reveal the experimental steps taken, the timing, and often the status during and outcome. Often this includes crucial environmental factors that can be readily measured with sensors. All of these data points are useful when the goal is to be able to replicate an experiment to validate results. It won't be long before these machines are integrating traces directly to digital lab notebooks and lab information management systems.
Throughput is the name of the game with automation. This is one of the best measurable metrics to tell the research value brought in by an automated system. How many samples can it process in a day? How does that compare to the daily amount of samples that the non-automated process tops out at? Typically, a functional automation system will be able to blow a small team of scientists out of the water when in comes to sheer throughput of sample processing. Robots don't get tired or confused as easily, they don't make silly mistakes based on fatigue or impatience.
Lean headcount is one of the many quantifiable benefits of a functional automation system, because the robots are capable of such a higher throughput than even a couple of humans, but those two humans could theoretically operate the robot around the clock to handle much higher workloads. This is another one of the obvious measurable metrics that show success or failure of an automated system.
Another crucial value is the high accuracy of these precise instruments to handle small volumes, within an error rate that rivals the best human pipettors. So the increased quantity of work throughput is actually coupled with higher accuracy and reliability on a validated instrument like this, so you are actually doing more work and also better work.
Barriers to embracing lab automation
Given that this is an extremely tech-oriented way of approaching traditional lab research, it’s not uncommon at all for lab automation to be intimidating to young and older scientists alike. Not long ago that certain demographics were struggling to catch on to techy concepts like email, social media, and video chat; crucial technologies in which illiteracy is a massive weakness in the work world today. General unfamiliarity with the technology is a huge barrier to bringing new automation into a lab.
This is because scientists are trained skeptics, and this is a good thing. I commonly hear scientists looking to find reasons to discredit automation technology, commonly by finding ways in which the automation could fail. It’s important to remember that the goal of the automation is not to handle every single lab task, but only those that are repetitive in nature, well tested, and repeatable. The bottom line is that it's easy to find cases in which automation will not as readily succeed, but that certainly doesn’t diminish the value of bringing this technology into the lab to do what it does best. Don’t be mislead by these apparent shortcomings in the technology, and don’t let your natural skepticism prevent you from adapting and growing with the industry.
Of course automation instruments can be expensive. Given this high cost and the cultural barriers associated, bringing in automation projects can be a tough sell to risk-averse management. Luckily, plenty of evidence exists out there to help tip the scales when building a case. The truth is that automation projects should be approached like any other information technology project. When executed properly, they provide the infrastructure to support efficient research operations to make scientists lives easier. However when mismanaged, automation projects can quickly resemble a dumpster fire fueled by cash. Take the right steps to plan and manage your automation projects, and the resulting value will be clear.
One of the biggest cultural obstacles to this technology is the fear of being replaced by lab automation, and this isn't unique to the the pharmaceutical industry. At first this appears to be is a logical fear because there are so many examples of how automation is taking away jobs. That may be the case here in some labs, but overall I would argue that this is creating jobs. Sadly I believe that this fear will not serve scientists well, nor will it not change the rightful role that these technologies have in the lab. The best defense to being phased out by this next wave of technological innovation is to embrace laboratory automation, not fear it; making the decision to acquire skills related to this tech is actually the perfect way to keep yourself from getting left behind.
The bigger picture: lowering the cost of drugs
So why is increasing pharmaceutical R&D efficiency so important? Drug discovery is an extremely expensive process, often said to cost at least half a billion dollars and 10 years. However bringing a new drug to market can cost upwards of billions. Most of this cost is in the discovery of new drug leads, so labs are finding clever ways to implement automated systems that allow them to increase research capabilities.
While the cost for bringing automation into the lab seems expensive, but it’s really hard to put a price on good science. Overtime, utilizing these machines in the way described above lead to more efficient research and can readily scale throughput. This actually makes science happen quicker, and can reduce the cost of drug R&D. High R&D costs for drugs relate directly to high pricing to protect that investment, so at the end of the day lab automation has the power to make life-saving drugs cheaper for patients in need.
Have anything to add? I'd love to hear your thoughts. Leave a comment to share.