Garbage In, Garbage Out

It’s all too easy to get excited by the launch of new gadgets and gizmos. Certainly, most chromatographers hanker after the latest cutting-edge hyphenated chromatography system that promises increased functionality, speed and sensitivity. But there is a time-tested adage: ‘garbage in, garbage out’ – and it’s probably more pertinent today than ever before, especially as the boundaries of performance are continually pushed. The bottom line? You can have the best chromatography system in the world, but if you inject poor quality and/or inconsistent samples, your results will be trash.

It is widely recognized that sample processing and preparation creates a bottleneck in lab workflows. It affects labs all over the world and accounts for around two thirds of the overall time spent on chromatographic analysis. Indeed, a 2015 survey established that 60 percent of chromatography users (in a range of industry sectors) believe the biggest challenge in sample prep is the time and labor intensity of the procedures required (1). The number of steps involved before a sample is ready to analyze can be extensive; just over half of those questioned regularly undertake three or more prep techniques per sample, and around 5 percent use seven or more.

But resources are not the only problem. The multifactorial nature of sample preparation means that it harbors potential for error at any one of its multiple steps. From poor sample storage to the addition of impure solvents and inconsistent dilution techniques, it’s no wonder sample collection, preparation and processing is by far the largest source of error in analytical laboratories (2).

Finally, as detectors get more and more sensitive, the susceptibility to sample quality and variability increases significantly. It stands to reason that any interventions that minimize error during preparation and ensure as consistent a sample as possible will have a positive impact on data quality. So, what can be done?

To reform sample quality, we must challenge the traditional view that pre-analytics are a series of separate processes. Despite the fact that many instrumental chromatographic techniques have matured (and automation of some kind is now relatively commonplace), I believe there is much more that can be done to address the remaining sample prep logjam. To date, automation has tended to take two separate paths; namely, robotic sample preparation or automated sample injection. Whilst there can be advantages to automating these two processes in isolation, the entire sample preparation workflow still includes numerous manual steps, most notably the need to transfer samples from one system to another. This ‘loophole’ in the system consumes analyst time and leaves the door wide open for potential errors to creep in.

With the emergence of fully integrated automated systems that combine sample preparation (for example, standard addition, liquid/liquid extraction, SPE) and injection in a single platform, the intermediate error-prone, time-consuming manual steps can be eliminated, improving accuracy and repeatability. Moreover, because such systems can run 24/7 without human intervention, such ‘smart’ automated systems can facilitate high sample throughput, allowing scientists to focus on the skilled analysis and interpretation of the results rather than on time-consuming wet chemistry or sample injection.

For example, we recently worked on a metabolomics study of algae cell cultures using a PAL RTC platform, which involved a pretty complex pre-analytical process (3). We combined Bligh and Dyer (LLE) extraction with dual-column UHPLC-MS/MS separation and detection in an automated setup (which included adding the set reagents and splitting the aqueous and organic fractions prior to injection into the UHPLC system). Even with an initial manual step, the automated method proved significantly less labor-intensive than the manual technique and also allowed the results to be directly subjected to a library search using LC-MS/MS data in SWATH mode (SCIEX), meaning that further targeted experiments to identify unknowns were unnecessary. Moreover, when compared with the manual method, the automated setup had better repeatability for both aqueous and organic fractions.

Analytical scientists, regardless of industry or sector, are all aiming towards a common goal: achieving the best quality results as efficiently as possible. Taking a smart approach to automating the sample preparation process cuts the ‘garbage’ being fed into the sample analysis process and improves data quality and speed of delivery to researchers – and thereby reduces the ‘garbage’ output from the lab.