Vendor Column: Improving Laboratory Operations with Standardized Informatics Solutions

Before my current role as a Laboratory Informatics Senior Marketing Manager, I was a LC/MS laboratory manager. I distinctly recall my first realization as a laboratory manager that I didn’t always have a good sense of the workload on the analytical instruments. In my monthly report to our group director, I would detail the injections run on each instrument. And for the most part, capturing this information was a manual process because all of the instruments were controlled by isolated workstations. That is, all data was collected onto the hard drive of the local PC attached to instrument. Of course, there was an automated tool that would copy the files to a network file server, but copying would periodically fail, requiring a manual cross-check several times a week. Although the files were centrally located on a file server, the full meta-data was captured on the workstations, meaning that any metrics on instrument usage had to come from individual workstations.

Ultimately, I wrote a Visual Basic .NET utility to query all of these individual workstations to capture the injection details. But, before I wrote the utility, the information gathering would take about eight hours to capture and compile manually. After implementation of the tool, the time required for this process was reduced to about five minutes. To accomplish this, a number of these fixes were put in place to overcome the fact that data was being captured on the hard drive of local workstations. Really what was needed was client-server configuration where all data automatically was captured to a relational database housed in a secure data center.

Although I was a lab manager in pharmaceutical discovery, a standardized client-server architecture for LC/MS made sense for a number of reasons. For example, with a centralized data repository, I could easily locate data for medicinal chemists whenever they needed help interpreting or reprocessing this information. In the workstation and file server architecture, reprocessing data always meant manually locating and downloading files to a different PC (different from the one the data was acquired on), but with a client-server configuration, reprocessing could be performed directly on the server without recopying data files. The medicinal chemistry organization had also adopted an electronic laboratory notebook, so automatically linking experimental results to notebook entries would be much easier in a client-server configuration.

However, as a laboratory manager, one of the biggest advantages a centralized system afforded me was the ability to look at instrument usage dash boards. For example, I could look at the LC/MS method usage by project. I would often find that a new project had been started by a medicinal chemistry disease group that required the creation of new LC/MS methods. Without this information, my staff would be caught off guard if the medicinal chemists were suddenly complaining that the instrument was no longer working for them. I could also better plan routine maintenance with more timely information and schedule maintenance for the least busy times of day, e.g., cleaning an electrospray source. And during the budgeting cycle, I would be better positioned to request capital improvements because I could demonstrate instrument usage and need.

In the business world, having dashboard style information is common and typically referred to as business intelligence (BI). BI makes sense for business managers and it also makes sense for laboratory managers; both can manage their respective operations more effectively. In the past, this type of information has largely been acquired by having laboratory staff manually enter usage information into a spreadsheet (or build widgets as described above). However, if you have been following this column, you know that many laboratory based organizations are working to standardize their chromatography data system (CDS), scientific data management system (SDMS), electronic laboratory notebook (ELN) and laboratory information management system (LIMS) onto a single vendor’s platform, which can be deployed globally. Since all of these solutions utilize a relational database, BI should become more common for use by laboratory managers.

Article by Chris Stumpf, Waters Corporation

• Read more about Chris Stumpf and our other columnists >>>

The views represented in this article are solely those of the author and do not necessarily represent those of John Wiley and Sons, Ltd.