Bringing bioinformatics to the bench: SubtiWiki 5.0 and new web tools for research
Doctoral thesis
Date of Examination:2025-10-17
Date of issue:2025-10-29
Advisor:Prof. Dr. Jörg Stülke
Referee:Prof. Dr. Jörg Stülke
Referee:Dr. Johannes Söding
Referee:Prof. Dr. Burkhard Morgenstern
Referee:Prof. Dr. Gregor Bucher
Referee:Dr. Heiko Liesegang
Referee:Prof. Dr. Kai Tittmann
Files in this item
Name:Thesis_Christoph_Elfmann.pdf
Size:22.2Mb
Format:PDF
Abstract
English
Model organisms are central to biological research, as their study provides crucial insights into fundamental processes of life and enables biotechnological applications. The intensive research on these organisms spans multiple disciplines from biochemical and genetic studies to large-scale omics analyses and computational predictions. To organize this wealth of information and to aid scientists in the development of novel research hypotheses, model organism databases integrate and present data in a comprehensive manner. Bacillus subtilis, the model organism for Gram-positive bacteria, is the most studied prokaryote next to Escherichia coli. SubtiWiki is a B. subtilis database that was first released in 2009. It contains a large body of expert-curated content, datasets from relevant studies, and interactive features for the intuitive exploration of data. As new data and tools are being released, continuous improvement of the framework is necessary to stay up to date with ongoing research. Recently, the introduction of AlphaFold and related tools has revolutionized biological research, as the accurate prediction of protein structures has opened up a world of structure-based analysis. In this work, SubtiWiki has received a completely new foundation (version 5.0), which builds upon modern, state-of-the-art frameworks and includes many new features, such as an interactive display for gene neighborhood conservation, integration of metabolite data, a mutant fitness graph, and a 3D protein structure viewer. AlphaFold and related resources offer great opportunities but are not always easily accessible to lab scientists. To keep up with these latest developments in computational structural biology, Foldseek, a structure similarity analysis tool, was integrated, which allows identification of remote evolutionary relationships between proteins that cannot be detected by sequence comparison alone. Moreover, the results of a proteome-level B. subtilis study were incorporated, where protein-protein interactions were experimentally identified and computationally modelled with AlphaFold-Multimer. To facilitate the evaluation of such predictions, the PAE Viewer was developed. This web tool allows for the assessment of multimeric AlphaFold predictions and also supports experimental validation through the display of crosslinks. In addition, the framework of SubtiWiki was applied to two additional bacteria to provide the respective scientific communities with extensive resources for research: MycoWiki for Mycoplasma pneumoniae, and ListiWiki for Listeria monocytogenes. With these changes, SubtiWiki provides the B. subtilis community with substantially improved utility and new tools to approach research questions. The cutting-edge features incorporated into the platform may help to bring the latest advances in computational structural biology to the lab.
Keywords: bioinformatics; subtiwiki; bacillus; subtilis; mycowiki; listiwiki; pae viewer; model organism; web tool; database; protein structure; alphafold
