Computational methods for the analysis of bacterial gene regulation
|PhD ceremony:||Mr R.W.W. Brouwer|
|When:||January 10, 2014|
|Supervisor:||prof. dr. O.P. (Oscar P) Kuipers|
|Co-supervisor:||dr. S.A.F.T. van Hijum|
|Where:||Academy building RUG|
In this research, we checked the effectiveness of previously developed operon prediction methods and we developed new methods for the prediction of operons. A total of 29 operon prediction methods were reviewed of which only 4 had implementations which were freely available. From these methods, the most commonly used indicators for operons were distilled. With these features we developed our own operon predictions. These predictions are based on minimal feature sets combined with various machine learning methods to yield accurate fast-learning cross-organism operon predictors. In the second part of this research, gene expression during the growth of L. lactis MG1363 was monitored using DNA microarrays. Trends in known processes were analyzed and new actors in these processes were identified. This resulted in a genetic network which can aid in the formulation of testable hypotheses and associated experiments.