INRA UMR SPO Montpellier/ Centre de Recherches de Biochimie Macromoléculaire (CRBM) - Centre National de la Recherche Scientifique (CNRS)
University of Edinburgh Medical School, Division of pathway medecine
Stage-onderwerp 2012-2013: Systematic analysis and visualisation of microarray studies
Download Presentatie (1.95 MB)
Samenvatting eindwerk 2012-2013: Microarray analysis and visualisation of a compendium of pathogen induced gene expression studies
Gene expression studies using microarray technology generates huge amounts of data. The ability to process, analyse and visualise this data in a meaningful way is still a great challenge.
In this thesis, I have implemented processing methods, using the statistical software package R to process, analyse and visualise microarray data. For the studies, raw data, sourced from both public (GEO) or private (GPX) repositories, was first loaded into R. It was then normalised and processed. This step was necessary to make the data comparable across arrays. The next step was to use this normalised data to generate graphs, each graph containing expression of one gene across a number of different conditions or treatments. The purpose of these graphs was to be uploaded onto GEVisE, a website for exploring gene expression profiles from microarray studies through these graphs. Therefore, a thumbnail had to be made for each graph as well. This was done using an image manipulation program called ImageMagick.
In the second part of the project some of these datasets were used to test a hypothesis regarding the genes involved in the sterol pathway. The intention was to test if the genes of the sterol pathway were significantly expressed when a treatment was applied. Using the Fisher’s exact test, the significance of their expression values was calculated. The outcome of the Fisher’s test provided us with useful information for future projects.