Scientific pursuits in R&D and QC laboratories over the last decades have been greatly assisted by the introduction and use of new and increasingly sophisticated instruments, techniques, and technologies. This is good news, but this increase in capability has also created ever increasing amounts of laboratory data and information. More data and information have driven new discoveries and better quality measures and control, however, today’s laboratories and scientific organizations have become virtually inundated with data and information. Analyzing, managing, integrating, and storing this data and information has become a major challenge and the volume is just going to keep increasing.
To the rescue come a variety of laboratory data management, analysis, visualization, reporting, and storage systems. Most instruments come with their own data and analysis systems and the alphabet soup of laboratory data management systems keeps growing (LIMS, ELN, LES, SDMS, etc.). Likewise, sophisticated data analysis, mining, reporting, and visualization tools have proliferated.
So problem solved, right?
Unfortunately, all of these solutions and systems have actually made the problem more complicated as now it is feasible that the data and information resides in multiple places in multiple systems both commercial systems and homegrown systems that may or may not “talk” to each other.
Two Environments, One Solution
We have found that the scientific data and information challenge described above resides in two distinctly different environments. The first environment is the “Greenfield” and the second environment is the “Amoeba”.
The Greenfield– This environment often occurs in scientific organizations like biotech startups that have had significant growth and therefore their laboratory data management and sharing needs have also grown. Being nimble organizations, they will have addressed this challenge by implementing point solutions and even creating some of their own software systems. Unfortunately, nothing “speaks” to anything else, and without that integration, their critical laboratory information has become difficult to access and share. The scientists in these organizations are spending more time working on data management software solutions than doing their real job – pursuing science.
The Amoeba– This environment generally exists in large, multi-site scientific organizations. These are often Fortune 100 companies and may be international as well. These organizations will have virtually every laboratory data management, analysis, mining, reporting, and visualization tool ever created, as well as several that they have made themselves. Sometimes this type of environment just evolves and sometimes it gets generated through a merger, acquisition, or major reorganization/centralization laboratories or scientific functions. Just like in the Greenfield environment, their laboratory data management and sharing needs have grown, but although they have tons of data systems, nothing speaks to anything else and critical laboratory information is difficult to find, access, and share.
The Solution: Scientific Data Plan
Whether your environment is the “Greenfield” or the “Amoeba”, the way to face these challenges is by creating a Scientific Data Plan, executing it, and then revisiting it on a regular basis to make adjustments for your evolving needs. Developing a scientific data plan is a multi-step process.
Step 1 begins by establishing your “Starting Point” or “Current State”. This is basically documenting your laboratory data generating systems (i.e. the instrument systems), the laboratory data management systems (i.e. LIMS, ELN, LES, SDMS, etc.), and the laboratory data analysis, mining, reporting, and visualization systems. Additionally, you need to map both your laboratory workflow and your laboratory data flow. Another key aspect to examine is the IT infrastructure that is supporting your laboratory systems.
Step 2 is to determine what will be your “Goal Point” or “Future State”. This is your opportunity to optimize workflow and data flow, enable systems integration and information sharing, remove redundancies, establish and push standardization and generally streamline and empower your entire research, discovery, or quality endeavors. You need to document which laboratory data systems need to “talk” to each other as well as “talk” to other business and enterprise systems. Your future state should also address leading edge IT technologies such as mobile devices and cloud computing as well as any data and information security challenges that these may raise.
Once you have your Starting Point (“A”) and Goal Point (“B”) established, your Scientific Data Plan is how you are going to get from “A” to “B”. Sounds simple but in reality developing the plan and the associated strategy can be quite complex. There are many ways to go and prioritization and a phased approach will be critical to your success. And don’t forget, your laboratory data needs are always changing so periodically dust off your Scientific Data Strategy and Plan and reevaluate. You will be glad that you did.
“Greenfield” or “Amoeba” – How have you addressed your laboratory data and information challenge? Did you develop a Scientific Data Plan and implement it? How has this improved your workflow, effectiveness, and efficiency? Do you revisit your laboratory data strategy and plan on a regular basis?