|
Computers in Chemistry STANDARDS FOR CHROMATOGRAPHY DATA SYSTEMSASTM adopts protocols for analytical data interchange (Andi)Terrance A. Rooney
 Does that same reasoning apply to the chromatography data system (CDS) marketplace? Would customers buy and use more CDSs and be more proficient if the systems were built around standards? Well, the evidence over the past dozen years is that although chromatography standards are not absolutely essential for a market to grow, there are certainly benefits to be gained by being able to use a common format, at least for doing chromatography data exchange. Sometimes the use of standards comes about because of demand from customers, which is largely the way the standards for mass spectrometry (MS) have evolved. The members of the American Society of Mass Spectrometry have been involved to ensure that vendors' software is able to exchange MS data. With CDSs, the desire for a standard was driven largely because the people involved with developing it thought it would benefit the users. But most users were not demanding it from their preferred vendor (and for the most part, are not demanding it even today). In fact, occasionally there was some reluctance to standardize on the part of some vendors' marketing representatives because the establishment of a standard would make it much easier and more convenient to switch from one vendor to another. In the spreadsheet, word processing, and database software industries, there are several examples of one company overtaking another partly because files created by the competitor could be read. In the spreadsheet market, Lotus 1-2-3 once had more than an 80 percent market share, so any other company had to read Lotus files in order to compete. In the word processing arena, the early lead enjoyed by WordPerfect also made it the format that others had to read to entice users to switch.
ESTABLISHING PROTOCOLS
The history of establishing a standard protocol for chromatography began in 1987 with an American Society for Testing and Materials (ASTM) Committee working during the Pittsburgh Conference to see whether there was enough common ground upon which to build a standard. The head of the committee was Dave Nelson, a veteran of the analytical instrument industry who had worked for Varian and Hewlett-Packard before leaving the latter in 1979 to found, with partner Harmon Brown, Nelson Analytical (later purchased by Perkin-Elmer and now known as PE Nelson). When Nelson started the company, he was convinced that desktop computers (terminology used before IBM made "personal computer" a household word) would play an ever-increasing role in CDSs and analytical instrumentation. Nelson had seen the role standards played in the enormous growth of the PC industry, and he thought the chromatography market might benefit from having a standard data format. But things were not as simple as that. Nelson, now retired to the California coast in Santa Cruz, recalls that the ASTM meeting at the 1987 Pittsburgh Conference was somewhat frustrating, and the establishment of a standard appeared in his words to be "going nowhere" because of the many disparate viewpoints among the users and the vendors. When Nelson left the meeting, he ran into Dieter Hoehn, at that time the head of Hewlett-Packard's Analytical Products Group and a board member of the Analytical Instrument Association (AIA). Nelson and Hoehn agreed that it would be worthwhile to seek the backing of the AIA in hopes of more fully involving the chromatography instrument companies in the development of a standard. In 1988, the AIA board of directors approved the establishment of a task force, headed by Nelson, to recommend a mechanism by which CDSs could exchange data. Mike Duff, the executive director of AIA (now known as Analytical Life Science Systems Association, or ALSSA), recalls that the goal was to make a "plug and play" standard kit that would make implementing the standard straightforward for vendors and users. The budget established by AIA for the development of the CDS standard was $50,000, with member companies donating the time and programming skills of software engineers to hammer out the protocols. At first the AIA thought it would be able to charge for the standards, and thereby underwrite some of the cost associated with the project. But companies objected to this idea, and later the kit was sold for a nominal amount to companies that wanted to implement it.
ANDI PROTOCOLS
What is included in the Andi format for the ".cdf " file? It comprises header information, such as the operator name, instrument used, column and detector descriptions, plus the raw or processed data. The benefits of having an exchange mechanism for chromatography data are fairly substantial from the user's viewpoint. One of the most important uses of the common format is that it allows laboratories that have data systems from multiple vendors to compare chromatographic data collected on one system with that collected on another. Methods developers for large national and multinational companies might use it to check on the consistency of chromatographic results when the equipment and software in two locations are not identical. For example, an analysis done in Europe using a Waters Millennium software package can be compared directly with one done in the US using the PE Nelson Turbochrom software. Integration parameters for the two systems can be tuned and optimized so that comparable results are achieved in both locations without either location having to change existing equipment.
Another major use of the Andi protocols is that they allow users to archive their data to just one format, which is independent of the computer architecture, operating system, or version of the chromatography software used. Several major chemical and pharmaceutical manufacturers are now using the Andi protocols as the preferred format for long-term backups, although most companies still do backups in the data systems' native format as well. The use of the Andi format for backup allows users to change vendors for their CDS without the fear that stored data will no longer be available for inspection. Of course, some users will always prefer to rely mainly on hard copy output from integrators and printers for archiving because those data can always be read. In fact, the use of inkjet technology in chromatography integrators was driven mainly by the desire for permanent storage of results; the thermal paper used previously faded badly with time. The long, tortuous history of trying to establish a standard protocol for CDSs has been filled with a significant amount of work by many people and many companies. Nelson believes that from a technical standpoint most of the early work has remained unchanged. From a marketing standpoint, the relatively limited budget did not allow the message about the existence of the standards to be widely disseminated. A number of articles on the Andi protocols were published, and talks at various meetings were presented, but for most chromatography users, the question was "Andi who?".
IMPLEMENTING STANDARDS
On advice from AIA members, the task group looked into ASTM as a forum for further standards activities. The benefits provided by ASTM include a framework for the generation, management, and distribution of standards, and, most importantly, the requirement for the participation of end users as well as vendors. Duff believes one reason the standards have not been more widely used is that the committee was composed of software vendors and did not include end users. By contrast, most ASTM committees are a combination of vendors and users. The recommendation to transfer standards activity from AIA to ASTM was presented at the 1997 spring ALSSA board of directors meeting. The board approved the initiative, and the task group then focused its energies on establishing an ASTM subcommittee dedicated to analytical data interchange. Working with ASTM representatives, the group proposed a subcommittee to ASTM Committee E49 on Computerized Systems and Chemical and Material Information. The inaugural meeting of Subcommittee E49.08 on Analytical Data Interchange Protocols was held 8 July 1997 at ASTM headquarters in West Conshohocken, PA. The meeting was hosted by Teresa Cendrowska of ASTM, who explained ASTM structure and how ASTM standards were managed. Working groups were established for the chromatography standard and MS standard, and the members of the subcommittee agreed on a plan to move existing standards into the ASTM framework first and then to pursue new standards activities. The standards were reformatted in ASTM standard format and sent out for review. Subsequently, the standards went to ballot to receive the required votes from members of the E49 committee. At the next subcommittee meeting in San Diego on 10 November 1997, the group addressed comments that had been submitted with returned ballots. The input from the voters was incorporated into the standards, and the standards were sent to ballot once more at the end of last year.
THE RESULTS The subcommittee will address new issues such as updating the existing standards, working to make the standards consistent across techniques, addressing how to distribute software utilities to support the standards, and considering new standards activities. Subcommittee E49.08 for Analytical Data Interchange Protocols has about forty voting members so far. The makeup of the subcommittee includes a strong vendor contingent, but, in this reincarnation of the CDS standards effort, end-user participation should increase, and the standards will then reflect the needs of the marketplace. For information on joining this activity, contact Teresa Cendrowska at ASTM (610-832-9500) or view the ASTM Web site ( www.astm.org). For information about what software products support the Andi standards, contact your chromatography or MS vendor. For many users, the ability to save all their chromatography data in a standard format adds to the security and manageability of the laboratory. For labs with software from multiple vendors, the Andi format also ensures that data exchange and comparison will be relatively painless. And for CDS consumers, Andi means that switching from one vendor to another can be done with some degree of confidence that older data will still be available for emergency situations, which happen all too often.
SEE OTHER HOT ARTICLES FROM THE MAY ISSUE: |
