Metabolomics analysis of plants, pets, pests and pathogens

Metabolomics is at the centre of several major projects involving plants, microbes and parasites in the institute involving at least a third of current staff supported by more than 3.5 million pounds in grants. From a strategic perspective, a major aim of the proposal is to improve sample throughout, data precision and reproducibility and ensure a longer-term data `shelf life¿ in existing programmes. At the same time the upgrading of existing equipment will allow us to promote new research interactions in the institute as a whole. One important added value in the project will be the opportunity to interact with BBSRC-funded staff in the Aberystwyth Metabolomics Centre who are developing data models and a database strategy for the UK metabolomics community. Thus the data structures derived from these instrument upgrades will be analysed by Experimental Officer-level staff specialised in advanced data analysis and database development and supported in future implementations of the ArMET database maintained by the MetRO consortium as part of the Plant and Microbes Metabolomics Initiative. Similarly, the higher quality data will also be available for research designed to improve data analysis routines. From a technical perspective the present proposal is based primarily on extending our capacity to automate the production of reproducible, high through-put direct infusion ESI-MS metabolite fingerprint data and GC-tof-MS metabolite profile data by the application of NanoMate injection technology for LC-MS applications and use of Automatic Liner Exchange coupled with Thermal Desorption (TDS) sample injection technology for GC-MS applications. Both of these sample delivery systems greatly improve precision in terms of signal purity and by favourably altering the signal to noise ratio. In GC-MS this allows the reproducible deconvolution of a much greater number of metabolite peaks, providing greatly increased metabolome coverage. In addition, in specific configurations, the TDS system will allow a more comprehensive analysis of metabolites by virtue of access to more volatile chemistry. Furthermore, ionisation reproducibility and signal intensity in both instruments is also improved which allows the more robust annotation of less intense signals, allowing their inclusion in pre-processed data tables. Finally, both sample presentation methods will avoid sample `carry-over¿ which causes major cross-contamination and contributes to instrument drift problems in larger experiments. These latter factors often lead to `batch specific¿ characteristics which can confound particularly multivariate analysis of these high dimensional datasets. By minimising such factors it may be possible to combine data derived many months apart in future analysis without major problems associated with batch-specific variance which will provide a great boost for a metabolomics database and data analysis strategy.