A highly scalable system for genome based computational microbial diagnostics

In-silico primer and probe design based on nucleic acid genome sequence data is of great interest in many areas of research and economy concerning microbial diagnostics, i.e. molecular identification and detection of bacteria and viruses. Currently available software tools have reached a performance limit due to the extremely fast growth of publicly available sequence data of microorganisms produced by high throughput sequencing technology. The aim of this thesis has been to develop a highly scalable integrated system for microbial in-silico diagnostics based on microbial genome sequence data. The first central component is a framework for efficient retrieval and management of sequence and annotation data from different heterogeneous data sources with the ability to obtain data sets ranging from small subsets to all available microbial genomes. To support fast non-heuristic oligonucleotide string matching and applications for primer/probe design on this data, required by computational microbial diagnostics, a new memory independent nucleic acid sequence index structure is developed as the second central component. In order to handle the huge amount of available and impending data, central components of the integrated system have been optimized and parallelized to efficiently utilize multi-core-architectures and high performance computers as both will continue to multiply the number of computing cores to increase computing power in the future. The integrated system will allow a large amount of users to continue their work on desktop computers while also enabling new HPC-based applications and research not possible so far with the currently available software.