Design, Development And Implementation Of Bioinformatics Tools For The Mining Of Microarray Gene Expression Data

The aspiration to understand the molecular complexities of the human body to the highest level in an efficient manner creates unique problems for the scientist in the design and implementation offunctional genomic studies. Furthermore with the involvement many hundreds ofgenes in even the simplest biological processes, how does one begin to identify the genes from the entire genome that playa significant role within a particular biological phenomena. The answer lies with microarray technology providing a resourceful solution enabling the simultaneous isolation ofgenes that may participate in any biological process under investigation, from an entire genome in a single experiment. Although an incredibly powerful technology, with the ability to generate vast amounts ofgene expression data, the technology and generated data is futile without the use of bioinformatics tools data interpretation in a manner that is meaningful to a biologist. As biologists are not experts in computer science, the role of the bioinformatician is not simply the development of novel algorithms but to ful(v integrate them within applications that are user-friendly for a biologist to utilise without the inten'ention ofa computer scientist. In light ofthis, we have designed and developed practical applications for a biologist specific for functional annotation, gene chip fabrication and cross comparison ofmicroarray data. Underlying every microarray experiment is a specific biological question centred on Underlying every microarray experiment is a specific biological question centred on an area ofexpertise. While it is useful to view the functions ofsignificantly differentially expressed genes across multiple un-related disciplines, it is more important to understand gene function within the specialised field for a specific biological question in order to continue further focussed research. In light of this we have generated an application called the multifunctional Immune Ontologiser for the functional classification ofgenes and gene functions more suited for the molecular immunology expert. Due to this specific nature of some microarray experiments initiated to answer highly focussed biological questions, often, available gene chips may not be ofbenefit. In this instance, the use ofcustomised gene chips would be more beneficial. Thus we have provided biologists with a tool to extract biological informationfrom gene-sets to allow the customised creation ofany number ofgene chips. We have exploited this tool to create an oncology- and an immuno-tolerance gene chip for our research purposes. Lastly, with the public availability of several hundred microarray experiments, public repositories contain much hidden biological knowledge that has the potential to be highly valuable ifmined in the correct way. Thus we have developed MaXlab, the first fully functional application for the meta-analysis of biological signatures from Aflymetrix and cDNA microarray studies to gain further insights into related biological phenomena.EThOS - Electronic Theses Online ServiceGBUnited Kingdo