by Kristan Bahten, Sr. Product Marketing Manager, Thermo Fisher Scientific
It’s little surprise that proteomics is a fast-expanding sector. As, perhaps, the most valuable of the “omics,” it characterizes the expression of the genome, and provides insight into every protein modification and complex interaction. This usefulness is backed up in the finances too. Proteomics is on track to be worth $50 billion by 2027 with a projected annual growth rate of 12.2% from 2020 to 2027.1
Proteomics is not just a tool for the future, it is already used to diagnose disease, evaluate its progression and assess the efficacy of drug discoveries, and its usefulness only increases as we move towards the era of personalized medicine. Yet, despite the clear benefits and growth in this important field, proteomics is still relying on highly manual, time-consuming and costly processes within the sample preparation stages. These cumbersome steps are creating a bottleneck in analytical laboratories, diverting highly skilled scientists to carry out repetitive tasks, and increasing error and failure rates in the analysis stages.
The future landscape for proteomics must be a fully automated one, starting with sample preparation and running into the separation, identification and analytical steps that follow. This article explains why this must happen and how the industry might go about it.
Breaking down an art to create a science
The explosive growth in the proteomics industry has been possible because of recent advances in technology – namely the development of highly sensitive and automated liquid chromatography coupled with mass spectrometry (LC-MS). Scientists can now analyze thousands of peptides in a single run, delivering fast and highly accurate results.
Yet, despite the efficiency of the technology, the experimental design and sample preparation stages are anything but efficient. These stages rely on highly experienced scientists developing bespoke and manual methods, from scratch, for every analysis, using unique techniques and advanced chemistry to craft sample preparation.
This manual process consists of many stages, including cell lysis, reduction/alkylation, digestion, labeling, pulling, clean-up and peptide assays, and each step is susceptible to a range of issues.
First, sample preparation requires manual pipetting and labeling, followed by a process of tracking and transferring samples that differs according to the individual. Scientists may process many hundreds of tests in a day, making it hard to keep track of each sample. Add in the variations from each scientist’s individual techniques, and it can be hard for experiments to be reliably reproduced by other experts, even those working with the same workflows in the same laboratory.
Errors can easily creep into these manual processes and, since an error won’t be evident until the whole analytical process is finished, many expensive and valuable samples can be misprocessed and lost this way. If diagnosis or drug development is dependent on the analysis, these errors can have costly consequences both at a monetary and human level.
Secondly, valuable time and knowledge are wasted on manual tasks. Each step involves wait times and, end-to-end, the full sample preparation time can take a working day; when wait times increase, so too does the margin for error. When highly qualified scientists are repeating laborious and repetitive activities, they are not being deployed to value-added work and this can cause demotivation, which ultimately leads to workforce attrition. Not utilizing scientists for highly skilled and creative work that requires their knowledge and higher-order thinking is a waste of time, money and knowledge.
To fully democratize proteomics, its access must be widened to many more laboratories, not just the few that can perform the highly crafted rituals of sample preparation. To ensure this highly effective analytical tool is available to more organizations, the art of manual sample preparation must be automated into a standardized science. This is the only way to streamline proteomics research, increase analytical laboratory throughput, and ensure standardized and reproducible results.
The future is automated and it's closer than you think
As we move towards laboratory 4.0, embracing all that technological advances and digitalization have to offer, automation becomes an expectation and then a necessity. Many methods and techniques are already fully automated, so why not proteomics?
Of course, many parts of the proteomics workflow are fully automated, from LC-MS runs to the backend statistical analysis. The sample preparation stage is where the manual workarounds persist. Despite some sample preparation systems being marketed as automated, they still require many manual steps, including reagent dilution, transfer stages or complex statistical protocols. Understandably, scientists and laboratory managers have been wary about adopting these semi-automated methods that simply mask new manual steps. Many prefer to stick to their tried and tested methods, hopeful that truly automated systems may soon exist.
When automated sample preparation does come, it will need to be fully integrated with slick and streamlined downstream workflows, so that automation becomes part of the end-to-end proteomics journey. We see this already in many other analysis techniques, such as clinical diagnostics, and these can form a template for proteomics workflows.
Sample preparation automation will come, that is inevitable, but when it does, what should it look like?
Standardized
The process for sample preparation is already well-defined; set steps need to occur in a particular order and this lends itself well to automation. The challenge is around taking universal chemistry that performs well with every sample type and standardizing that for reproducibility in every laboratory setting. Essentially, wherever there is a chance for human intervention, and, therefore, human error, that step should be automated.
Intelligent
Even before sample preparation begins, the experiment must be fully conceptualized, and this means robust experimental design. Traditionally, consulting statisticians are brought in to help define the design and ensure the analysis answers the biological question. Advanced statistical tools should form part of sample preparation automation, linking the parameters of the analysis, such as the type and number of samples, and the values and factors that affect them, to the outputs required.
This approach will ensure that results are more meaningful and intrinsically tied to the ultimate requirement of answering the biological question. More advanced systems could recognize sub-optimal set parameters and alert the user to adjust the design, thereby, ensuring greater confidence in the results.
Democratized
A key goal of automation is to make analysis techniques available to a wider range of organizations. By automating sample preparation and experiment design tasks that would otherwise require highly experienced scientists or statisticians, the whole sample preparation process can be democratized and available to more laboratories. The full power of proteomics can, therefore, be utilized by more laboratories, allowing for greater insight across the industry, and, ultimately, furthering better and faster diagnostics and drug development.
Truly automated and integrated
The semi-automated systems of today prevent widespread adoption. A truly automated system will have no manual steps and be supported by pre-validated, kit-based reagents that require no additional mixing or formulation. Automated sample preparation equipment, reagents and software must also be fully integrated with hardware, software and consumables used in other areas of the proteomics workflow. Each element of the proteomics process will, therefore, be intrinsically linked, from the parameters set during sample preparation to the final peptide analysis stages. The true statistical power of the experiment design can then be linked to the data analysis at the end of the process and the biological question can be fully addressed.
Automated sample preparation will unleash the full power of proteomics
By removing the error-prone bottleneck of manual sample preparation, proteomics can be fully democratized, unleashing its full power and making it accessible to every organization that needs it. This will be the springboard to reducing errors in proteomic analysis, and the human and monetary costs that are associated with sample processing mistakes.
By decreasing hands-on tasks and reducing waiting times, we move toward walk-away solutions, processing costs are further reduced and scientists are freed to carry out value-added activities. By automating steps, through pre-validated, kit-based reagents, the time needed to prepare samples also decreases and the bottlenecks that prevent true analytic efficiency are widened. This enables more samples to be processed each day and expands the scope of proteomics analysis to more diagnoses and drug development opportunities.
Furthermore, introducing automated reproducibility helps to democratize the entire proteomics workflow, allowing this valuable analysis tool to expand into more and more laboratories.
Although full automation isn’t yet available in experiment design and sample preparation stages, it is surely on the horizon. Proteomics is too valuable a tool to wait.
References
1. https://www.alliedmarketresearch.com/proteomics-market#:~:text=The%20global%20proteomics%20market%20was,of%20protein%20structure%20and%20function
About the Author: Kristan Bahten is currently Senior Product Marketing Manager at Thermo Fisher Scientific in San Jose. In his primary role, he is the Product Manager for automated sample preparation systems for proteomics. Prior to his current role, Kristan was Product Manager for IC Systems at Thermo Fisher Scientific, concentrating on the Dionex ion chromatography products and was responsible for the successful launch of several new IC platforms and autosamplers. With a Master of Business Administration and a BS in Forensic Science and Chemistry, Mr. Bahten spent a number of years in the semiconductor, pharmaceutical, and environmental industries before becoming the Business Unit Manager for the Analytical Instruments division at PECO, Inc.