Derek Van Dyk

Proteomic analysis of dog tears for potential cancer markers.

The first reference map of the proteome of pooled normal dog tears was created using 2-dimensional polyacrylamide gel electrophoresis and the identity of a number of the major species determined using matrix-assisted laser desorption time of flight mass spectrometry (MALDI-TOF) and peptide mass fingerprint matching on protein sequence databases. In order to understand the changes in protein expression in the tear film of dogs with cancer, tears from such animals were similarly examined. A number of differences were found between the tears of healthy dogs and the dogs with cancer. Differences were found in levels of actin and albumin and in an unidentified protein which may be analogous to human lacryglobulin. These findings suggest that it may be possible to develop tear film analysis to provide a simple non-invasive test for the diagnosis and/or management of canine cancers.

The microbial proteome database--an automated laboratory catalogue for monitoring protein expression in bacteria.

Laboratories devoted to high‐throughput characterisation of purified proteins arrayed via two‐dimensional (2‐D) gel electrophoresis face an arduous task in maintaining a centralised and constantly evolving record of information relating to the characterisation of proteins and their responses following biological challenges. The Microbial Proteome Database (MPD) has been conceived as an in‐house resource for complementing the plethora of genomic databases available for such organisms. The database utilises commercially available software to provide an electronic ‘lab book’ of information obtained daily from 2‐D electrophoresis gels, image analysis packages, protein characterisation methodologies, and biological experimentation. The MPD begins from a single 2‐D gel image (a 2‐D ‘reference map’) with clickable spots that link to a ‘protein catalogue’ (ProtCat) with spot information including protein identity, changes in expression determined under experimental conditions, cellular location, mass, and pI. The entry for each protein then contains further links to gel images corresponding to the presence of the particular protein within different subproteomes (as defined by the pH of narrow‐ and wide‐range immobilised pH gradients or from differential extraction methods used to determine the location of the protein within a functional cell). The database currently contains information from strains of three microbial species (Escherichia coli, Pseudomonas aeruginosa and Staphylococcus aureus) and 32 master gel images. The rapid accessibility of information obtained from microbial proteomes is an essential step towards the integrated analysis of these organisms at the gene, transcript, protein and functional levels and will aid in reducing turnaround times between sample preparation and the discovery of molecules of biological significance.

Djinn Lite: a tool for customised gene transcript modelling, annotation-data enrichment and exploration

BackgroundThere is an ever increasing rate of data made available on genetic variation, transcriptomes and proteomes. Similarly, a growing variety of bioinformatic programs are becoming available from many diverse sources, designed to identify a myriad of sequence patterns considered to have potential biological importance within inter-genic regions, genes, transcripts, and proteins. However, biologists require easy to use, uncomplicated tools to integrate this information, visualise and print gene annotations. Integrating this information usually requires considerable informatics skills, and comprehensive knowledge of the data format to make full use of this information. Tools are needed to explore gene model variants by allowing users the ability to create alternative transcript models using novel combinations of exons not necessarily represented in current database deposits of mRNA/cDNA sequences.ResultsDjinn Lite is designed to be an intuitive program for storing and visually exploring of custom annotations relating to a eukaryotic gene sequence and its modelled gene products. In particular, it is helpful in developing hypothesis regarding alternate splicing of transcripts by allowing the construction of model transcripts and inspection of their resulting translations. It facilitates the ability to view a gene and its gene products in one synchronised graphical view, allowing one to drill down into sequence related data. Colour highlighting of selected sequences and added annotations further supports exploration, visualisation of sequence regions and motifs known or predicted to be biologically significant.ConclusionGene annotating remains an ongoing and challengingtask that will continue as gene structures, gene transcription repertoires, disease loci, protein products and their interactions become moreprecisely defined. Djinn Lite offers an accessible interface to help accumulate, enrich, and individualise sequence annotations relating to a gene, its transcripts and translations. The mechanism of transcript definition and creation, and subsequent navigation and exploration of features, are very intuitive and demand only a short learning curve. Ultimately, Djinn Lite can form the basis for providing valuable clues to plan new experiments, providing storage of sequences and annotations for dedication to customised projects. The application is appropriate for Windows 98-ME-2000-XP-2003 operating systems.