by Rodney J. Nash, Ph.D., Principal Scientific Officer, Omni Homogenizer Portfolio, and Alexandra Blancke Soares, Ph.D., Sr. Application Specialist, Automated Robotic Solutions at Revvity
When the Human Genome Project concluded in the early 2000s, it offered a boost to personalized medicine and helped quickly grow the repertoire of multiomics tools. Labs suddenly had access to and a need to integrate assays that could probe complex biological systems from many different angles. Labs were tasked with maintaining the integrity and quality of their data outputs while juggling higher sample throughput and sustaining the continuous implementation of multiple prep and analytical methods.
Tissue sample analysis, for example, has impacted our ability to evaluate pathophysiology from basic to clinical science disciplines while promoting numerous diagnostic and therapeutic advances.1,2 Despite these advances to sample prep, many processes and initiatives continue to experience challenges when it comes to reliable and reproducible sample preparation methodologies.3-5 Robust sample preparation methods are the preliminary and often one of the most important phases in workflows, preparing a key sample for downstream analysis. Most high-throughput labs stake their reputation on reproducibility. Resilience is key, in handling fluctuations in sample traffic or staff turnaround. These labs often adapt and see opportunities that require smart and modular scalability, and this same mindset applies when on-boarding multiomics. While approaches may differ based on lab priorities, our experiences note that the core competencies within pre-processing, analytical processing, and post-processing are typically shared.
Pre-processing with priority
Within the theme of tissue sample preparation, historically, it has been laborious and tedious work that relied on techniques like cutting, grinding, smashing, and filtering samples into a partitioned section of a tissue for further analysis.1,2 Each sample poses its unique challenges, even before we speak of analytes. In a workflow, even small inconsistencies or batch variability in this pre-processing step can snowball into systemic issues or chronic failure modes making any statistical analysis irrelevant.
The challenges that come from each sample speaks to the importance of proper sample prep. For example, cultured cells, free-floating analytes (such as cfDNA or protein from a preserved blood draw) would not undergo the same process optimization as with tissue. Tissue disruption requires multiple protocols that should be optimized to a biospecimen’s physical nature such as toughness, consistency, size, and weight. Tissues are arguably an important sample matrix in many biomarker studies, offering a window into the environmental niche. Proper early sample prep offers a complement to the success of downstream technologies; and these combinations of technologies are needed when studying samples such as tumors.6 Analysis of tumors benefits from not only thorough processing of tissues that generate cell lysate, but also from gentler dissociation for single-cell or intact-nuclei preservation for multiomic assessment that leverages some outstanding sequencing techniques of today.
Mitigating risk
Once your "pre-prep processing" is set up, the true testament of its durability is observed by the analyte-specific workstreams it feeds in parallel. This is where many inconsistencies might be observed such as inconsistent buffy coat collection or uneven tissue shearing, which in turn can impact the accuracy or tightness of your resulting data. Therefore, sample preparation plays a critical role in procuring and preparing targeted analytes for accurate downstream analysis.
Lab automation, which improves standardization, analytical reproducibility, and throughput, is fundamentally changing multiomic workflows. Automation tools, such as automated next-generation sequencing (NGS) library prep systems with pre-validated methods or robotic autosamplers for liquid chromatography and mass spectrometry, are becoming a new standard. While these are very powerful standalone tools, we see an increasing demand for robotic integration of these tools that combines them with pre-analytical steps.
Automating partial or full workflows results in freedom from worrying about avoidable errors, like pipetting mistakes, sample loss or mix-up, and sample variability caused by slight protocol deviations between personnel. Instead of focusing on improving human factors like manual dexterity, labs can now spend time fine-tuning workflows to new levels of efficiency, giving them the high-quality results they need for their research. While in-run specs like fragment size consistency or adapter ligation efficiency or excessive signal spikes offer valuable QC insights, monitoring longitudinal consistency can help identify and avoid quality issues in the final sample analysis. Controls for batch effect and reagent lot variability are commonplace, but modern automation enables active monitoring for a large quantity of variables beyond that. For automated liquid handling workstations, it is crucial that they can detect not only outright errors, but even the subtle changes such as changes in dispensing pressure and retention time between steps. Technologies like pressure-based liquid level detection (LLD) play a vital role in avoiding timely and costly errors like running out of reagent mid-assay without warning.
Connecting your liquid handler with other instruments through robotic integration opens up even more possibilities for tracking and analyzing process data. Combined with a powerful orchestration software, the timing of individual workflow steps, e.g. centrifugation, incubation, and reagent addition, can be optimized and tweaked to maximize throughput or maximize reproducibility. Measurement results from intermediate analysis steps can be analyzed mid-workflow and the results automatically applied to downstream processes like sample normalization.
Now, what’s exciting for us is that automation challenges the ‘more-is-better’ mindset. Robotic precision in pre-analysis steps combined with machine learning analysis algorithms are helping labs reduce sample consumption without losing statistical confidence, which is critical for growth of data sets and harmonization of multiomic interpretation. While costs associated with consumables and reagents cannot be eliminated, automation enables labs to ‘do more with less’. Automated liquid handling facilitates assay miniaturization that would be either extremely strenuous or impossible to achieve by manual pipetting. Instruments integrated into automated workflows can be used to full capacity even at night or on weekends as they require only minimal or no human intervention. Considering some trends observed in demographic changes, labs might soon face a shortage of qualified personnel. If this were to happen, lab automation can help both mitigate the effects of a shrinking workforce and enable researchers to focus more on what is relevant, interpreting data and developing new ideas. Several case studies have shown an increase in productivity and throughput enabled by lab automation.7-9
While total lab automation might sound appealing as it seems to promise a complete and worry-free solution, we tend to recommend what some may call a less radical approach to lab automation. Since the late 1990s, laboratory management communities have spoken about how robotic ‘building blocks’ may be more attainable as compared to an immediate foray into total lab automation (TLA).9,10 This calculated approach will make more logical and financial sense to most labs by offering several advantages in incorporating scalable workcells.10,11
Adopting a modular, incremental approach to automation allows labs to scale capabilities as workflow needs evolve, providing a prudent and flexible path. Indeed, labs could start by automating their most critical bottlenecks and most time-consuming or error-prone steps and expand over time.
For labs looking to successfully implement lab automation, we’ve listed some of the core takeaways from our experience:
- Change management: Changing the current status quo can be challenging. Involve staff in the early phases of planning for lab automation and be open about risks and benefits. The goal is not to replace people with automation, but to help them become more productive and focus on tasks they enjoy.
- Data Integration: Many labs use LIMS to collate data and analyses from different sources. While connecting all data sources with a LIMS through an API can be considered, a simple communication through file exchanges between the LIMS and automated workstation significantly reduces both implementation time and future requirements for support as it is more vendor independent. This foresight is critical, as it ensures seamless integration even with multi-vendor instruments as your lab's needs evolve. It will likely help reduce the risk of future software and command compatibility issues, making it far easier to add new functionalities without requiring a complete system overhaul.
- Strategically Utilize Lab Space: A modular approach lets you optimize your existing lab footprint. Find an automation partner that offers stackable or space-saving modules to design a dynamic layout where possible that could expand vertically or horizontally without a full redesign.
- Master Workflow Orchestration: Start by automating simple, high-impact tasks. As you add more modules, you can incrementally build expertise in dynamic scheduling and manage unattended runs.
- Progress Toward "True Walk-Away" Automation or keep it modular: Each addition of a new module can bring you closer to a fully automated workflow. This step-by-step approach allows for careful validation and optimization, leading to a robust, reliable system that reduces manual intervention and increases throughput. Another option is to keep individual workflows separated and contained within independent modules. While this approach might require more manual interaction, it offers more flexibility and simplifies error handling. For high-throughput and high-stake projects, planning with redundancies reduces the risk of complete project failures.
In conclusion, a modular strategy transforms potential complex challenges into manageable steps, enabling a more agile and efficient research environment.
Beyond post-processing
Planning and implementing process optimization in a lab with automation can be even more challenging when adding on the analytical requirements needed for multiomics, including data analysis, interpretation, and storage. Powerful bioinformatic capabilities are needed, along with solutions that address the challenges of data-banking in cloud-based or localized networks. Our bioinformatics colleagues who breathe digital network and code would agree that this is undoubtedly one of the immense post-processing challenges of this age that labs face today.12-15 It’s no longer a luxury, but a necessity as it brings all the work from the wet lab into actual translational applicability. When you add multiomics, such as proteins, your sequences, your epigenetic regulation, and your metabolites into your scope of work, you are navigating databases and accredited research data networks that house disease signatures, and consolidating it all into meaning. For labs with limited bioinformatics expertise or infrastructure, modular tools or modular access to tools can help reduce the barrier of entry for automation of dataflow of multiple parallel pipelines and cloud-based scalability of complex, integrated analysis.
Final thoughts
Given the aforementioned, automation can help bridge gaps that will increase efficiency and overall health of the lab and its people. All labs won’t face the same kind of gap or the same access to resources. Some are at the stage of lacking computing power, others need help with configurable pipelines or advice on what to start with, and still others may prioritize the ability to pivot that matches the culture of startups or new CROs. As with the democratization of information, so too has automation become less rigid, more modular to fit, and more openly utilized in supporting scientific rigor and quality. A good integration offers unified processes, harmonized protocols, and robust QCs to address the challenges we see in research today, such as execution of cross-study comparisons across continents or the exhaustion of irrecuperable limited samples.
Partnering with sample prep experts from industry can bring palpable benefits. The ability to leverage both scientific and engineering expertise collected from years of practical experience within a single organization is a key consideration for successful robotic integration and ultimately lab efficiency. Having supported researchers with new technologies for well over 50 years, it’s interesting to see how, as with many things in life, simplicity attracts complexity. Just as science evolves thanks to each discovery and technological milestone, automation can simplify sample prep and ease workflow scalability such that researchers can focus on challenging the new complexities we are now enjoying in the space of multiomics. As you consider automating your multiomic workflows, bring these considerations to your lab.
About the authors
Rodney J. Nash, Ph.D., is a Principal Scientific Officer for the Omni homogenizer portfolio at Revvity. He has an extensive background in Molecular & Cellular Biology, particularly in the development of cellular models for human disease. He offers domain expertise in primary cell cultures, diagnostic testing, and method development for cell dissociation. He is also a professor at Georgia State University and holds committee positions at several organizations, where he enjoys mentoring the next generation of scientists.
Alexandra Blancke Soares, Ph.D., is a Senior Application Specialist at Revvity, working with the Automated Robotic Solutions team, which brings over 25 years of experience in custom lab automation. With a background in molecular parasitology and cancer research, she helps translate manual lab protocols into automated workflows and designs workstations tailored to each lab's specific needs. Her goal is to make lab automation accessible across diverse laboratory environments, expanding its application beyond traditional high-throughput screening into mainstream lab operations.