Project Description

The field of proteomics involves the study of the entire set of proteins in a biological sample. High throughput techniques have been developed [8, 10] which utilise protein separation via two dimensional polyacylamide gel electrophoresis (2D-PAGE) [6] ,and protein identification using mass spectrometry. These techniques generate large datasets containing diverse types of data, for which standard formats do not currently exist. The types of data include: large images of two dimensional gels, the raw data from mass spectrometry, the results of analyses on the data and annotation. A format will be developed that can integrate these diverse data types into a formal structure enabling efficient data management and search facilities. XML (Extensible markup language) is an emerging standard for storing biological data in a structured format [4]. A preliminary study has been carried out to investigate the use of XML for storing experimental data from 2D-PAGE experiments [9].The study concluded that XML would allow the semantics of the data to be maintained and would provide flexibility with regards to the types of data that can be stored. However, significant further research is required to develop a formal schema for describing proteomics experiments. Research has been undertaken in the related field of gene expression experiments using microarrays. A document type definition (DTD) was described, which defined the format of XML to store the protocol and results of a microarray experiment [1]. One of the initial aims of the project will be to develop a standard format for proteomics experiments, such as a DTD or XML Schema [2]. The schema must provide a structure that allows easy integration of the experimental data with other resources, such as pre-exisitng protein and genome databases. An interface will be developed allowing biological researchers to store and view the experimental results with dynamic links to external databases. The images produced from 2D-PAGE experiments must be securely archived, and integrated with the experimental data, ensuring that specific protein data is linked to the correct position on the image. Software exists for loading XML documents into a standard tree structure that allows the data at individual nodes to be accessed [3]. However, visualing and editing the data requires either: a user experienced in XML structure, or software