Now that the need for high sampling efficiency has been met in spectroscopic instrument and software design, the inevitable consequence is the production of an incredible volume of information. Raw sample data, qualitative and quantitative measurement results, and pass/fail ratings are just some of the information generated, and this information comes from a range of analytical techniques with different instrumentation and users. Laboratory information management system (LIMS) soft-ware manages sample information ranging from scheduling to test results throughout its lifetime in the lab.

The strengths of LIMS tend to lie with techniques such as chromatography and non-instrumental wet chemistry, which are heavily used in the pharmaceutical arena for the verification of formulations and raw materials. Chromatography in these environments is an "analyzer" technique because the result is a set of quantitative answers or numbers rather than the data itself - ideal output for placing into LIMS. Molecular spectroscopy however has traditionally been more data-focused, and therefore not well suited for LIMS interfacing.

Spectroscopic analyses and results need to be coordinated with other analytical output. The growing acceptance of fast spectroscopy systems into these environments has outpaced the development of information management tools. Only recently have analytical groups begun to make efforts to link the output from spectrometers to the overall flow of the lab and to cross-technique sample management through systems such as LIMS. Also, with the recent development of spectroscopic analyzers that primarily produce quantitative results, their successful implementation in a QC or process environment necessitates interfacing output with LIMS.

The recent integration of two data and information management products, eRecordManager™ and Nautilus™ LIMS, both from Thermo LabSystems, Beverly, MA, and two molecular spectroscopy software packages, OMNIC® and RESULT® from Thermo Nicolet, Madison, WI, helps labs and quality testing groups to address these issues.

A pharmaceutical analytical lab typically is responsible for a broad range of materials testing and analysis. The analyses performed in a lab range from drug development feasibility to quality control (QC) for clinical study release to method development for production QC outside the lab. Different testing functions in the lab require different sets of test protocols so one type of sample may be analyzed using only one instrumental technique, while another may be analyzed in aliquots by multiple techniques. The output of such a lab will be a combination of both quantitative and qualitative information, with the final answer ranging from a spectrum itself to a quality rating based on a series of quantitative treatments of the raw data.

A large number of the instruments in this lab will have networked computers associated with them for data work-up. The LIMS system would reside on the network and would be used for scheduling certain types of sampling and for managing information about the sample throughout its lifetime in the lab.

In most laboratories, HPLC and gas chromatographs are well integrated with the local LIMS. If a lab is like most, the Fourier transform infrared (FTIR) spectrometer is not connected to the LIMS. The LIMS is used to tie all the results from different analytical methods for a given sample together. To do this with spectroscopic results, measurements might be noted in a physical notebook, then manually entered into the LIMS at a different workstation.

Suppose upon review of the sample results in the LIMS, a reviewer wants to see the spectroscopic data. In the common implementation described above, one must look up the lab notebook number in the LIMS, find the notebook, and look up the experiment record for the data. In the note-book, a spectral data file name and date might be recorded. The next step is to then find the actual data, which may be archived on CD in a storage facility. Once retrieved, the data must be taken to a data station on which OMNIC software is installed for examination.

Besides the complexity of record entry and retrieval in this manual process, a problem with this sort of 'unconnected' data archival is that it does not provide for ready searching or retrieval for the researcher with a frequent need to look for spectra of compounds. This inability to effectively manage the body of analytical knowledge in the lab is a costly problem. Even more costly is the possibility that the next breakthrough drug may have already passed undiscovered through the lab at some point and now exists only as a set of data archived in warehouse somewhere.

In a real situation, this rather complicated scenario would only be a small part of the picture, given the number of users, services, and techniques co-existing in the lab. Implementing the right set of tools will begin to solve the above issues. Management and archival of data that allows linking, searching, and quick, application-independent retrieval can be made simple with the implementation of eRecordManager software. The eRecordManager data archival system provides a central repository for data and metadata in the laboratory. In this laboratory’s implementation of the system, eRecordManager's archival engine could be set up to automatically sweep the appropriate data directories at scheduled intervals for instrument output (data). It will store the spectral data along with any relevant metadata, and create a complete and accurate XML-based copy of the record (data + meta-data) that can be viewed and manipulated without the need for OMNIC software.

The XML-based record is written in a file format called Generalized Analytical Markup Language, which is available in the public domain. This allows review of all records within a standard, maintainable interface that does not require the presence of the software that created the original data, regardless of technique, software, or instrumentation. This eliminates concerns about data retrieval in years to come. eRecordManager creates the accurate copy for approximately 150 different instrumental data types past and present based on the XML standard, potentially saving tens of thousands of dollars in maintaining computers and software used to create the original data. eRecordManager is capable of reading most analytical file formats in existence today, so data from all instrumentation can be connected into the archival system, including that from obsolete systems.

The eRecordManager database could be integrated with the LIMS database to allow the LIMS user to retrieve, at a single mouse click, all archived spectral data associated with any numeric or textual LIMS result. Measurements or other results based on data from OMNIC or RESULT can now be easily mapped into Nautilus LIMS. Furthermore, the analytical data from different techniques used on a single sample can be linked and retrieved into a single viewer. The impracticality of having all relevant software packages immediately at hand would no longer be an issue.

The maintenance and accessibility of an analytical database also becomes a realistic goal of the laboratory. The existing data converters and the ability to automate archival would make the future management of analytical knowledge go smoothly and quickly.

* This article originally appeared in the July 2002 issue of R&D magazine.