Automating fatty acid analyses from seeds: from field samples to data bases

Abstract

A goal of oilseed breeding is to improve oil quality by modifying fatty acid profiles in crops such as soybeans and sunflowers. We have used robotic automation and process engineering to improve sample throughput and efficiency of fatty acid analyses to support breeding efforts. A custom integrated system was developed using a Perkin-Elmer Nelson automated chromatography data acquisition system, in-house relational data bases, and sample preparation robotics. Using field/plot data base information, breeders request oil analyses, generate laboratory sample numbers and bar codes, and fill custom sample trays with seed for shipment to the centralized analytical laboratory. In the laboratory, oil is extracted in the sample tray by a manual procedure and a custom robotic system adds solvent, pipettes the sample into vials, adds a derivatizing agent, crimp-caps the vials and loads them into autosampler racks for placement on the chromatographs. The system was designed to handle 2100 samples in a normal 8-hour day. Custom software is used to download sample information from a data base and generate sequence information for Perkin-Elmer Nelson's TurboChrom system which manages seven dual-injector Hewlett-Packard 5890 gas chromatographs. A Novell network server provides multiple user access to the raw data. Quality control of the chromatographic data is performed by ChromVue, a custom system of programs to display chromatograms for review. Final results are uploaded to the original field data base where they can be accessed and reported by the breeders.