Environmental Metabolomics Information System

Overview

The Environmental Metabolomics Information System (EMIS) provides an online capability for recording and sharing information about metabolites discovered in samples of plants and soil via the Shimadzu gas chromatography-mass spectrometry (GC-MS) instrument. The EMIS moves a largely paper-based and manual system to a more up-to-date, computerised approach for recording, summation, enquiry, and management of the plant and soil metabolome data. This enables handling greater data volumes, centralising all experiment data in one place, and improving the ability to search and share the entire set of historical data as well as perform detailed analysis.

Background

Metabolomics studies the metabolites present in an organism. Metabolites are chemicals produced by the process of metabolism (e.g. sucrose). Across the plant kingdom, there are around 200,000 different metabolites and they may have quite complex chemical structures.

Previous Situation

The raw mass spectrometry data is generated by a Shimadzu QP2010MS gas-chromatograph/mass spectrometer. This instrument generates data in a proprietary format with instrument metadata, but can also save data in a common format, but without metadata.  Raw data was manually collated into summary spreadsheets (.xls or .csv).

Experiment metadata (experimental parameters and so on) was recorded in a paper lab book, including a written record of the name of the raw data files generated by the experiment. There are about a dozen different kinds of experiments that produce mass spectrometry data, and each has different kinds of metadata. Raw data files were manually uploaded from the instrument and stored on the University of Sydney’s Research Data Store.

Benefits

The following benefits have been delivered:
•    Easy data access to the public and other researchers through a Web interface. This enables researchers around the globe to access information already captured without having to repeat the experiments themselves, thus saving costs.
•    Data from earlier experiments can be captured in a central place, and integrated with new data.
•    Improved productivity in research activities via greater computerisation, resulting in easier data entry and search/enquiry of stored data, and less manual data entry. It is estimated that EMIS will save approximately 221 hours per year in data entry and data retrieval tasks, saving around $20,000 per year.
•    Improved data management processes, enabling handling of larger data sets, reducing risk of data loss, and providing a more robust solution. Approximately 20 to 30 files (out of 500 to 1000 annually) are lost or misplaced. With a computerised database and adequate data administration procedures, it is estimated that such data loss will be eliminated.

Project Features

Collection descriptions will be fed to Research Data Australia as part of the University of Sydney’s ANDS-funded “Seeding the Commons” activities scheduled for 2011.

The EMIS increases the productivity of researchers by requiring less manual effort, and enabling access to the public and other researchers, consequently improving collaboration. Features include:

  • a utility to extract and export instrument metadata to be stored as part of the instrument run data;
  • a database for storing experiment definitions, experiment data, raw data files, and  processed summary data. This database will be backed up and secure;
  • a web-form for researchers to input details of experiments and their parameters, and associate this metadata to specific datasets, both raw and processed;
  • a website where approved researchers (collaborators) can upload legacy data, download data associated with experiments and input metadata;
  • a utility to automatically generate experiment summary data which provides a report on the analysed plant and soil samples and identifies the metabolites; and
  • an administrative facility which permits those nominated to set access privileges for researchers to data, add new metadata fields for metadata input, and associate metadata fields with experiments.