Annual Conference 2025 of the Association for General and Applied Microbiology - Jahrestagung 2025 der Vereinigung für Allgemeine und Angewandte Mikrobiologie

Chair

Prof.

Microbial Identification in the Environment

FG Identification & systematics

Historically, bacterial identification has relied on phenotypic characteristics observed in pure culture, and phenotypic diagnostics remain a cornerstone of routine bacteriological monitoring, particularly in hygiene assessments. At the same time, classic environmental microbiology put a particular focus on bacteria in their respective biogeochemical roles as drivers of element cycles or biotransformation processes, whereas taxonomic identification of environmental microbes remained conditional to prior isolation. The need to differentiate bacterial genotypes directly in field samples probably emerged from the study of environmental pathogens but has gained widespread adoption with rRNA-based phylogenetics: Barcoding by next generation sequencing has revolutionized microbial community ecology, and fluorescence in situ hybridisation with rRNA-targeted probes allowed for environmental quantification of genotypically defined pro- and eukaryotic populations while preserving aspects of their phenotype. However, phylogenetic resolution based on rRNA genes often proved insufficient for the distinction of ecologically unique genotypes within microbiomes. Genome-resolved metagenomics offers higher taxonomic resolution and additionally also provides insights into the "physiological potential" of individual populations. While invaluable, this approache also introduces a 'caterpillar - butterfly' dilemma when inferring phenotypes from genomic data. In my presentation, I will explore strategies for achieving sufficient genotypic resolution in microbial identification while integrating ecophysiological insights essential for ecological interpretation.

Short lecture

Raman spectroscopy as a comprehensive tool for microbial identification

Raman spectroscopy has become a powerful tool for the rapid, precise identification of microorganisms in diverse microbiological applications. Over recent years, our research has focused on developing predictive models to classify fungal and bacterial species using their unique Raman signatures, optimizing identification processes for environmental and clinical microbiology.A key challenge in microbial identification is differentiating similar species in complex environments. Initial studies on entomopathogenic fungi demonstrated that Raman microspectroscopy effectively distinguishes Metarhizium and Beauveria species in conidial form, achieving over 98% prediction accuracy across varying growth conditions. Expanding into bacteria, we developed models based on bulk Raman spectra of 21 species grown under standardized conditions, achieving high theoretical accuracies through machine learning, particularly support vector machines (SVM). Despite substrate differences, such as stainless steel versus silver, practical accuracy remained robust at ~80% when applied to independent spectra.We also addressed biofilm detection, focusing on dry-surface biofilms (DSBs) prevalent in healthcare settings. Raman spectroscopy successfully differentiated between vegetative cells and spores within these biofilms, advancing disinfection strategies. Incorporating convolutional neural networks (CNNs) further enhanced predictive models, especially for single-cell analysis. While bulk data models showed promise for single-cell predictions, challenges like fluorescence and photodegradation persist. This work contributes to creating adaptable Raman databases for microbial identification down to the single-cell level.Beyond applications, we investigated why Raman spectroscopy works so effectively. By analyzing spectral variations and baseline effects, we moved beyond peak analysis to biochemical interpretations, gaining insights into factors driving classification accuracy.Our research underscores Raman spectroscopy"s versatility in microbial systematics. From fungi in environmental studies to bacteria in biofilms and clinical diagnostics, this method offers a fast, reliable, and scalable solution for microbial identification, paving the way for broader adoption across disciplines.

Large-scale, community-based description of prokaryotes

A large fraction of prokaryotes remain poorly described, although genomes from isolates or metagenomes exist for many of them. This genomic information can be used to describe existing taxa and name new ones. The latter can now be done according to the rules of the SeqCode. Descriptions of taxa are essential to provide concise information to others unfamiliar with them, which is classically done in the form of protologues. As any single method of analysis provides only one piece of information at a time, accurate bacterial descriptions require multi-layered analysis to capture comprehensive information on the taxonomy, ecology and functions of the taxon under study. This requires sophisticated bioinformatic tools, which, combined with the large number of unknown taxa, makes the process of describing prokaryotes tedious and slow. Our bioinformatic tool Protologger (www.protologger.de) provides all the elements needed to describe novel prokaryote isolates or metagenomic-assembled genomes from their 16S rRNA gene and genome sequences within a few hours. It is hosted at de.NBI for stable deployment and accessible via a Galaxy interface for ease of use. Protologger has also been updated to use POCPu to determine genus delineation, analyse additional functional pathways, and include archaea. To speed up the process of prokaryote description by orders of magnitude, we propose a community-based approach. Protologger is being used to process all genomes available in the Genome Taxonomy Database (GTDB) (1,650 processed so far). All results are available via a Wikibase instance, allowing the community to contribute comprehensive descriptions. An API will be implemented to interface with the SeqCode registry, allowing the association of high-quality descriptions generated in this project with proposed name. Training videos have been produced to help users navigate the system and workshops will be organised by NFDI4Microbiota (https://nfdi4microbiota.de) to facilitate community participation.

FG Identification & systematics
Identification and classification of microorganisms

Appointment

Chair

Programme

FG Identification & systematicsIdentification and classification of microorganisms

Appointment

Chair

Programme

FG Identification & systematics
Identification and classification of microorganisms