Over the past decade, the biological laboratory has become an increasingly digitally enhanced place. These days a skills section of a CV is more likely to include various instruments and software systems than a list of molecular biology or cell culture techniques. It’s not just that the volume of data being captured electronically is greater than ever before; data are also being generated and diversifying at ever-increasing speeds. In a paper-based lab, those electronic data are simply printed out and pasted into an actual notebook for IP capture, commonly referred to by scientists as “arts and crafts time.”
With the advent of the Electronic Laboratory Notebook (ELN), this “arts and crafts” approach has been replaced by electronically copying and pasting or sometimes “printing” the files (Word, Excel, PowerPoint, GraphPad, DNA sequence files, etc.) as the mechanism for collating the data. This process typically has some electronic signature functionality, although often legal departments still demand that the ELN pages be printed and signed for, maintaining IP in the traditional way. At least in this manner, the content is searchable, however basic the search capabilities may be. It’s what we call a “sticker book” ELN.
But biologists want, need, and now demand far more than a sticker book. Their instruments are more sophisticated, their calculations and statistics are being performed by software applications, and they are looking for modern solutions that go above and beyond the entry-level sticker book capabilities. This is the 21st century, where we can ask our smartphones to find a sushi place nearby and instantaneously have a number of recommendations with other people’s reviews. Surely these lab systems should be integrated and automated to some degree. And, while we're at it, that coffeemaker is electronic, too, so why can’t it automatically have a hot espresso waiting for me when I finish this HPLC run?!
While no company has yet explored an integration to its favorite coffeemakers, there have been significant advances on the lab software side. There’s ELN, LIMS, SDMS, EBR, ERP (Electronic Laboratory Notebook, Laboratory Information Management System, Scientific Data Management System, Electronic Batch Records, and Enterprise Resource Planning). But as scientists, we just want our assay data to “auto-magically” be transferred to our notebook, where the results are calculated automatically, and those results are then sent to the scientists who submitted the samples. Can I just have a system to do that? This is the biggest question plaguing scientists and research and development (R&D) application support: What system will do what we need? The answer, unfortunately, is: It depends on what you do.
These days, the needs of biologists vary, depending on where in the R&D spectrum they lie. Early research needs to have the freedom to discover and require greater flexibility, while manufacturing needs process controls and minimization of deviations. Development lies somewhere in between; there is no solid line between flexibility and process control. It’s a gradient with the requirements overlapping between the various areas—research, development, and manufacturing.
Next-generation data management platform
Fortunately, systems are evolving beyond simply replacing paper, and next-generation data management platforms are emerging. These offer more than just a sticker book by enabling structured data capture and analysis as well as tight integration with various systems around the modern lab—the process. They are becoming the data and process hub for scientists rather than just another system that is being used in a collection of dead-end data silos. Some such systems can even be thought of as similar
to an “enterprise resource planning system for the laboratory of the future.” Forward-thinking organizations are making a shift toward data- and process-centric approaches rather than being document driven.
A next-generation data management platform is not the single solution to every biologist’s data and process woes, but it is an integral part of the ecosystem in today’s lab. My philosophy tends to be “use the right tool for the right job,” and this is imperative for efficient processes. You can create a spreadsheet to integrate an HPLC peak, but when Empower does such a great job of this out-of-the-box, why would you use Excel? Most scientists these days need some combination of experimental data capture, sample management, statistical analysis, and reporting. Modern biologics data management systems must handle all of these tasks in some manner, and exchange information with dedicated specific scientific tools when required. They must act as the hub for the user and not try to do everything.
Integration can mean different things to different people, but an old plate reader that runs on Windows 98 and only sends data to a dedicated printer is not going to have a lot of integration options. This is where figuring out the use case, documenting and optimizing the scientific “business process flow” is critical to getting the most out of the lab systems available. After all, research is only as good as the data it’s based on. Transcribing from instrument to paper, paper to Excel, and Excel to email/PowerPoint/Word reports leaves lots of room for error, and quality is guaranteed to be negatively impacted.
Handing off data in an automated manner is a great advance, but it is useless if those data cannot be accessed and shared in a collaborative manner. As much as we like to credit great discoveries to singular scientists, they are typically made by the combined efforts of a team of lab staff and the brilliance of the discovering scientists, e.g., Crick, Watson, Wilkins, Franklin, and their research student, Gosling. Collaboration is a major accelerator for discovery, and robust and flexible data and process management systems can support it by allowing the data to be effectively searched, enabling users to collaborate internally and externally using social tools such as tagging.
In addition, the data need to be high in context and minable in a meaningful manner, rather than a Google-like search for any document that contains “Green Fluorescent Protein.” Scientists are demanding solutions that allow cross-experimental data aggregation and reporting, trending and statistical analysis over and above basic text search capabilities. This enables them to ask questions of their data like, “How have the control values been trending in the past month on instrument XYZ?” as simply as they might ask their smartphone about sushi spots.
In today’s scientific data landscape, an effective data management platform is increasingly becoming the central hub for scientific data, and their use in biologics is growing rapidly. There is no single, out-of-the-box system that can handle all scientists’ needs because, of course, these requirements vary so much. However, applying solid business analysis up front does help justify and implement enterprise biologics lab data management and process systems. When properly implemented, these systems can have dramatic effects on lab throughput, quality, collaboration, and ultimately scientific innovation. By ensuring the foundations are right, they offer good data, in context with fast, easy access at the right time.
Jarrod Medeiros is Senior Consultant, IDBS, 25 Burlington Mall Rd., 1st Floor, Suite #107, Burlington, MA 01803, U.S.A.; tel.: 781-852-0019; e-mail: [email protected]; www.idbs.com.