Informatics Convergence Presents Opportunities and Challenges
Informatics Convergence Presents Opportunities and Challenges
The future of ELN will require innovative and disruptive thinking
It is no wonder that laboratory informatics users want the same level of convenience they experience in their lives outside of work. They do not want to have to manually transfer data between silo applications, nor is the thought of complex system integrations particularly exciting.
Users want a simplified experience that eliminates multiple interfaces and also consolidates platforms to reduce costs. According to our surveys, the majority of potential users would prefer to buy a single platform from one vendor if they could.
These consumer desires are pushing vendor solutions like laboratory information management systems (LIMS) and electronic laboratory notebooks (ELN) outside of their traditional boundaries (Figure 1). ELN products are adding LIMS-like characteristics while, at the same time, LIMS vendors are adding ELN components, making convergence a reality in the laboratory informatics market. The blurring of traditional lines can be a bit confusing, especially as people try to define these categories based on historical assumptions — particularly with ELN, which can be described many different ways. We see more and more RFPs being sent to vendors asking for both LIMS and ELN capabilities in a single solution. Does the supplier build everything? Do they partner, offer only their existing capabilities, or do they acquire?
|Figure 1: Expansion of capabilities blurring lines between informatics categories|
The benefits of ELN are significant and sales continue to grow strongly,1 but the future could take the technology down many different paths. Despite poor economic conditions, sales of products and services are projected to be $230 million in 2012, up from approximately $25 million in 2004.2 This is now nearly one third of LIMS sales. With this rapid growth, combined with the trend of convergence, ELN is at an interesting transition point in its evolution. ELN has gone from a minor to major informatics category in a very short period of time. As we look to the future, it is fitting to take a step back and look at the progression of convergence up to this point and examine some of the risks it presents going forward.
Historical perspective on convergence
It was a simpler time for ELN back in 2004, as there were clear distinctions not only between LIMS and ELN, but between categories of different ELN products; we coined two classifications “specific” and “non-specific.”3 Non-specific solutions were intellectual property protection systems, essentially content management solutions with extensions for security and electronic signatures. These “low science” systems were used mainly in research, focusing on capturing documents supporting patents, archiving and collaboration between researchers. Contrastingly, “specific” or “high science” products were application software solutions to address workflow and data analysis in areas such as medicinal chemistry and formulations.
|Figure 2: Three characteristics of software|
As many procedure execution systems interfaced directly to LIMS for transferring sample and result data, they were, in many ways, front-ends akin to a LIMS module — or at least one that should have existed in the first place. Sensing opportunity, procedure execution ELN vendors began to add tools for designing complex workflows, new instrument interfacing capabilities, and modules like metrology and analyst certification. Even limited sample management modules were added. As LIMS-like characteristics of procedure execution systems grew, some customers started to question if they even needed a LIMS at all, particularly if they intended to interface to an ERP. LIMS can be difficult to implement and maintain; compromising on the full functionality of LIMS could allow a simpler architecture to support.
Both the growth of procedure execution ELN and the prompting of vendors by LIMS customers to address the needs of the analyst have resulted in alliances and introductions. Thermo Fisher and ABI (now LabVantage) formed partnerships with companies like Symyx (now Accelrys) and Labtronics (now PerkinElmer). STARLIMS, LabWare and LabVantage introduced ELN modules based on their core LIMS product. In reality, these modules for execution of methods are extensions of LIMS capabilities, quite distinct from ELNs that are used in research.
In research, another phenomenon was occurring. Not only did content management solutions and “high science” systems merge capabilities, but the use of ELN expanded beyond early discovery research. In early discovery, workflows tend to be dynamic depending on the study being pursued. This generates significant volumes of unstructured data: documents, spreadsheets, images and data files. The fluid nature of the work does not easily lend itself — except for screening studies — to database representations of structured data. As early installations by research (particularly medicinal chemistry) showed management the substantial benefits of ELN, later-stage research and early development began to take an interest.
Later-stage research and early development posed new problems. These groups (e.g., pharmacology, drug metabolism, discovery pharmacokinetics) have a portion of their experimental work that is dynamic and variable in the type of data collected. However, they also have work that is highly structured, creating considerable volumes of structured data in processes such as in vitro drug metabolism and cell-based testing. Due to the diversity of the science, the need for study-specific flexibility, and ease in adaptation to new workflows, LIMS was never broadly accepted. This was compounded by a lack of desire by many companies to put in two solutions — ELN for variable tasks and LIMS for structured processes — because of the cost of maintenance and requirement to have a simpler user experience. This forced an acceleration of capability convergence. ELN products started to take on LIMS-like characteristics, such as structured data management, workflow templates, calculations and reporting.
It did not end there, as successful deployments further exposed the advantages of the technology across the enterprise. Later-stage development wanted to increase their productivity as well. Soon, systems began popping up in analytical development, bioprocess development and other late-stage groups, but the capabilities in ELN were not sufficient. Assay requesting, task assignment and sample tracking were required to support their operations — all very LIMS-like features. Consequently, companies such as Accelrys, Agilent, Edge, IDBS, PerkinElmer, Rescentris and Waters built some level of request, task and sample management into their ELN, though most prefer not to use the “L” word to describe these features.
Is convergence for the better?
It can be debated whether the blurring of lines between former discrete categories of informatics products is a good thing. Since procedure execution ELN is, for the most part, addressing capabilities that should be present in LIMS, and with many research ELN products taking on additional capabilities for sample management and task workflow, is ELN just becoming another LIMS or is it redefining the category? Is convergence bringing risks to be aware of?
|Figure 3: Technology adoption characteristics|
An example to illustrate this point is the evolution of LIMS. One of the challenges faced by LIMS vendors is their legacy; most of the major systems have their origins back in the 1980s and early 1990s. Back then, technology was a major differentiation with object orientation, use of Oracle forms, workflow engines, and use of the Web which came a bit later. As customers pushed monolithic systems to supplement their feature sets, it became increasingly difficult to change the underlying code base. Resultantly, you have, in many cases, complex systems that have not kept up with the changes in technology (e.g., trying to “Web-enable” a legacy product) and are not compatible with modern user interface designs.
According to our research, LIMS has not penetrated more than 50 percent of its total available market4 (in pharmaceutical QA/QC this number rises to 70 percent, driven by regulatory demands). One of the factors impacting sales is that the market entered the conservative phase a few years back. While pragmatists want all the features they can to support their needs, conservatives value simplicity. Many LIMS failed to make this transition, as current customers often place a higher value on new features and bug fixes, rather than any disruptive technology.
In his book, The Innovator’s Dilemma,5 Clayton Christensen points out that many great companies fail “by doing everything right” by continually focusing on existing customers and improving existing products by adding more features. Existing customers, therefore, become barriers to innovation and new thinking. Established companies with mature products become susceptible to disruptive technologies. They tend to ignore them until the market is established or big enough for them to jump in. In most cases, it is too late. As a parallel, LIMS companies ignored ELN for almost 10 years. As ELN has become an established category that poses risks to cannibalization of sales, it is interesting to see the sudden interest.
But — we are noticing the trend of convergence pushing ELN in a similar direction, stuffing more features into architectures that were designed for a different purpose. Not only have they acquired LIMS-like capabilities, several have added (or are adding) capabilities, such as raw data archiving, document management and inventory management. The risk is that the systems — which were originally very straightforward and simple to operate — will become increasingly complex to use and will follow the same path of stifling innovation.
This risk is especially true for systems used primarily in research. These products have document-centric designs based on supporting the former first-to-invent needs of the U.S. patent system. As the U.S. moves toward first-to-file with the passage of the America Invents Act, the document motif is both restrictive and archaic. Trying to combine “paper-on-glass” and structured LIMS-like workflows could lead to some pretty un-intuitive user experiences and open the door for disruptive technologies.
In the era of the iPad-savvy populus, a challenge for developers is to support the desires of convergence in a user experience that is inviting and pleasurable. The need for new capabilities should be supported in a modular, service-based architecture to avoid the mistakes of the unwieldy monolithic system. Balance is possible, but the highly competitive nature of the market and feature demands of potential clients make it a difficult challenge for any vendor.
As the masters of consumer convergence, Steve Jobs and his team at Apple proved that usability, technology and modular features (i.e., apps) can be combined into a desirable experience. While others failed, Apple showed that it takes holistic thinking, a strong emphasis on the user experience and capabilities that customers did not necessarily even tell them they wanted. As personal computing entered the conservative phase of the market, Apple realized that users wanted an engaging — and mobile — experience to expand the use of technology beyond pragmatists. Jobs said, “A lot of times, people don’t know what they want until you show it to them.”6 This is very true for ELN; many designs are based on traditional paper notebooks and spreadsheets — users’ current reality. Convergence will challenge these designs like nothing else. The future of ELN will require innovative and disruptive thinking to provide users with an engaging user experience far above traditional approaches.
1. Elliott, Michael H, “What are the Benefits of ELN?”, Scientific Computing, March 2010
2. Elliott, Michael H, Electronic Laboratory Notebooks: A Foundation for Scientific Knowledge Management Edition V, Atrium Research & Consulting, September 2011
3. Elliott, Michael H; “Are ELNs Really Notebooks?”, Scientific Computing, July 2004
4. Elliott, Michael H, Fourth Electronic Laboratory Notebook Survey, Atrium Research & Consulting, December 2010
5. Christensen, Clayton M. The Innovator’s Dilemma: When New Technologies Cause Great Firms to Fail, Boston, Mass., Harvard Business School Press, 1997
6. Business Week, May 1998
Michael Elliott is CEO of Atrium Research & Consulting. He may be reached at editor@ScientificComputing.com.