News

Continuation of the work in Halle (Saale)

Fri, 31 Jan 2025 16:24:00 +0100

Our group’s final project meeting on “Computational Systems Biology of Infections and Antimicrobial-Resistant Pathogens” at the German Center for Infection Research (DZIF) brought together students and researchers from Tübingen, Leipzig, Gießen, and Heidelberg in Halle (Saale).

Prof. Dr. Andreas Dräger (Martin-Luther-Universität Halle-Wittenberg) opened the event, officially launching the final meeting of the Deutsches Zentrum für Infektionsforschung (DZIF)-funded project. A highlight of the conference was the keynote lecture by Prof. Dr. Ralf Steuer (Universität Leipzig), who explained the concepts of resource-balance calculations in metabolic networks.

Student presentations covered a range of topics. Arthur Neumann (Eberhard Karls Universität Tübingen) introduced SBSCL, a Java™ implementation of various simulation algorithms for SBML. Dóra Viktória Molnár (University of Tübingen) presented the machine-learning method TFpredict, which identifies proteins with transcription factor activity in both eukaryotes and prokaryotes, together with Michael Gaas, who further explained the latest developments in TFpredict.

Additional presentations showcased our software tools ModelPolisher, refineGEMs, SPECIMEN, and NCMW (Dr. Reinahen Mostolizadeh), emphasizing their role in computationally modeling the microbiota of the respiratory tract, mainly focusing on ESKAPE pathogens. Dr. Nantia Leonidou (DKFZ Deutsches Krebsforschungszentrum) discussed metabolic models of the dangerous pathogen Acinetobacter baumannii. Dario Eltzner and Bahaa Ziadah (University of Tübingen) provided insights into the latest advancements in the model annotation tool ModelPolisher.

Gwendolyn O. Döbel, Carolin Brune, and Nina Roßbach (MLU) presented new methods for automatically generating high-quality genome-scale metabolic models.

With the conclusion of our DZIF-funded project, our affiliation with the University of Tübingen also officially ends. It was a special moment to bring together current and former group members, students, and invited researchers at the University of Halle to celebrate the successful completion of this project.

This working group is thus being merged into the Chair of Data Analytics and Bioinformatics in Halle, where numerous new tasks are on the agenda in addition to previous challenges. Further information on the new Chair of Data Analytics and Bioinformatics can be found here.

Second doctoral degree of this working group

Mon, 25 Nov 2024 18:00:00 +0100

Nantia Leonidou is the second doctoral student to successfully present her doctoral thesis in this working group, thereby completing her doctorate. The examiners were Prof. Dr. Andreas Dräger, Prof. Dr.-Ing. Oliver Kohlbacher, Prof. Dr. Nadine Ziemert and Prof. Dr. Hannes Link. Prof. Nathan Lewis, Ph.D., also participated as a member of the advisory committee (TAC). An important part of the work was carried out in the laboratory of Prof. Dr. Aurélie Crabbé and Prof. Tom Coenye at the University of Ghent, where Ms. Leonidou worked together with Lisa Ostyn. Important experimental tests of model predictions on A. baumannii were carried out by PD Dr. Monika Schütz at the University Hospital of Tübingen.

Algorithm for detecting and solving energy-generating cycles

Mon, 29 Apr 2024 12:45:00 +0200

Streptococcus sanguinis is a facultative anaerobic member of the Viridians Streptococcus group. The commensal is usually found in the human nasopharynx, where it is seen as an antagonist for pathogens. However, it can also colonize the gut and is considered one of the most common pathogens for infective endocarditis when it enters the bloodstream. Therefore, a deeper understanding of the metabolism of S. sanguinis can help to develop new control methods. Genome-scale metabolic models are a valuable tool for this. Based on the annotated genome sequence of an organism and the reactions and metabolites based on it, they offer the possibility of simulating the entire metabolism in the computer. In this way, such models can support experimental biomedical research. However, creating such models is a complex process that cannot yet be fully automated. For example, automatically generated metabolic models often contain energy-generating cycles contradicting introductory thermodynamics. In this work, Tobias Fehrenbach developed a tool that detects and removes such cycles. By applying this method, he created a strain-specific model of the opportunistic pathogenic bacterium Streptococcus sanguinis SK1 that meets the current systems biology standards.

Reconstruction of a metabolic model for Staphylococcus warneri AW25

Wed, 06 Mar 2024 10:45:00 +0100

S. warneri belongs to the genus Staphylococcus and is widespread on human skin and mucous membranes. The bacterium acts both as a commensal organism and a potential infection source. Infections are primarily observed in immunocompromised individuals and newborns. To research new treatment methods and better understand the bacterium’s metabolism, a model of the AW25 strain of S. warneri was created in this bachelor thesis. The validation of the model included identifying essential genes that provide information about weak points and potential drug targets. Sophia Krappen examined the growth of the bacterium in silico on various media, such as the synthetic nasal medium SNM3. A comparative analysis with the Staphylococcus aureus model iYS854 highlights standard and unique metabolic features. The generated model can serve as a starting point for further investigation of S. warneri and its interaction with other organisms.

SPECIMEN: A workflow for automated strain-specific metabolic modeling

Wed, 26 Jul 2023 10:00:00 +0200

In recent years, numerous pathogens, e.g., Klebsiella species (Klebsiella sp.), have been shown to develop resistance against antibiotics quickly. To combat these resistances, new targets and drugs for treatment need to be discovered. To assist wet lab experiments, computer models of the metabolic networks of the organism in question can be curated. However, detailed metabolic modeling to curate a high-quality model can take time due to the number of manual steps included.

In her master thesis, Carolin Brune introduces strain-specific metabolic modeling (SPECIMEN), a Python package with an automated pipeline as its core. The workflow facilitates automated, strain-specific model curation based on a high-quality template model. Additionally, the package includes functions for further extending, refining, and analyzing the curated model, reducing the manual curation needed to create a high-quality model.

The workflow has been successfully tested using a Klebsiella pneumoniae (K. pneumoniae) genome (strain MD01) and a high-quality template model from a different strain of the same species, producing a new model for Klebsiella pneumoniae MD01.

Reconstruction, validation, and phenotypic predictions strain-specific metabolic model of Staphylococcus capitis

Wed, 28 Jun 2023 12:17:00 +0200

Staphylococcus capitis is a Gram-positive bacterium belonging to the coagulase-negative staphylococcus (CoNS) family. On the one hand, it is considered to have a beneficial skin microbiota due to its ability to generate an acidic pH environment and thus inhibit the growth of pathogenic microorganisms. On the other hand, clinical interest in the bacterium is increasing because it has been identified as the cause of late infections in neonatal intensive care units. The presence of S. capitis was detected on medical surfaces as well as in the nasal cavity of medical staff. The bacterium has a unique property, biofilm formation. This ability enables its survival in unfavorable environmental conditions. As part of her bachelor’s thesis, “Analysis of metabolic phenotypes of Staphylococcus capitis using restriction-based modeling,” Ms. Anna Lefarova developed a strain-specific genome-level metabolic model (GEM) of the Staphylococcus capitis organism. The computer model was built following a highly standardized protocol for genome-scale reconstructions and validated against available laboratory data. In addition, the phenotypic predictions regarding growth on the different carbon and nitrogen sources were made. The generated model was compared against automatically generated models for S. capitis. The manually curated metabolic reconstruction provides a basis for the in silico simulations. It should help to deepen the understanding of the metabolic capabilities of S. capitis.

Strain-specific metabolic models of Corynebacterium striatum

Wed, 17 May 2023 16:45:00 +0200

Antibiotic-resistant bacteria are leading to more and more infections worldwide. According to a Hungarian study, a disease with the bacterium Corynebacterium striatum has been associated with the need for patient ventilation even more frequently since the COVID-19 pandemic. Yet it remains largely unexplored, and the mechanisms by which it moves from being a component of the average human microbiota to a pathogen are unknown.

Metabolic models can be used to gain more insight into its metabolic behavior and potential pathogenicity factors. In a project for two master theses, Famke Bäuerle created five strain-specific genome-scale metabolic models of C. striatum and validated them using laboratory experiments. Using two experimental methods, the models provided differentially good predictions on four different media (LB, RPMI, M9, and CGXII).

In parallel, the Python package refineGEMs (github.com/draeger-lab/refinegems) was developed for other researchers to examine and curate metabolic models.

The Python package and models have already been published on BioRxiv as part of a preprint.

Reconstruction, Comparison and Growth Simulation of Strain-specific Metabolic Computer Models of Staphylococcus haemolyticus

Tue, 16 May 2023 12:45:00 +0200

In times of increasingly spreading antibiotic resistance, the comprehensive investigation of opportunistic pathogens is becoming more and more critical. So-called in-sillico approaches (i.e., biological experiments carried out on the computer), such as the simulation of strain-specific metabolic models on the genome level (GEMs), make an essential contribution to investigating pathogens, as they enable the rapid testing of plausible laboratory scenarios. As part of her master’s thesis, “Reconstruction of strain-specific metabolic models of Staphylococcus haemolyticus,” Ms. Gwendolyn O. Gusak developed three new GEMs of the organism Staphylococcus haemolyticus (SH) and compared them to GEMs automatically generated for the same organism. Their new workflow from genome to draft GEM shows good results, as 98.69% of all protein sequences obtained valid identifiers from the NCBI database. Her resulting model iShIMITSC147GOG23 showed similar results compared to SH’s other models. In addition, as part of her work, she extended the script collection called refineGEMs to include an algorithm that fills gaps in the models, which she plans to continue working on after completing her thesis. Comparing the MEMOTE scores of the new models with the automatically generated models shows that the quality of the models is the same, except for the oldest model from 2017, which has a low MEMOTE score of 66.08%. The results of their growth simulations mostly show doubling times in a plausible range for laboratory research. She makes her work on the software and the models themselves publicly available on GitHub and in the BioModels database common to systems biology. After completing this work, Ms. Gusak plans to have her results peer-reviewed and published by a journal.

Successful disputation of our group member

Fri, 24 Mar 2023 16:30:00 +0100

We are pleased with our former group member, Thorsten Tiede, for successfully defending his Ph.D. thesis. He joined our group through a project funded by the US NIH (the National Institutes of Health) to contribute to developing the SBML standard specifically for JSBML. SBML also played a considerable role in his doctoral thesis, as he wanted to make this standard as valuable as possible for cancer research. To do so, he converted it into a graph-based data structure. We congratulate Dr. des. Thorsten Tiede, and wish him success and all the best for his future career and private life.

Targeted computer modelling to accelerate antiviral drug development

Tue, 21 Mar 2023 15:59:47 +0100

Effective drugs against viral diseases like COVID-19 are urgently needed now and in the future. The emergence of viral mutants and yet unknown viruses could push vaccines to their limits. The DZIF scientist and bioinformatician Andreas Dräger from the University of Tübingen is working on a computer-based method that can help to accelerate the time-consuming identification and development of antiviral agents. Using a novel analysis technique that applies to any virus and host cell type, the research team around Dräger has now created a model to detect additional host cell targets that allow inhibiting SARS-CoV-2 replication.

“Efficient pandemic preparedness requires new, broadly effective antiviral drugs against which the viruses cannot quickly develop resistance,” explains Andreas Dräger, junior professor at the Univer-sity of Tübingen and member of the Tübingen Cluster of Excellence “Controlling Microbes to Fight Infections – CMFI”. “But drug development takes too much precious time, which is urgently needed in an emergency.”

Dräger wants to remedy this situation through computer modelling. In 2021, the Tübingen research group identified a human enzyme in the model—guanylate kinase 1—indispensable for virus replication and can be switched off without damaging the cell. Now, the bioinformatician has developed a new model with his colleagues to examine the effectiveness of their targets. “Through an improved analysis technique, we can now specifically model viral infection in many different types of tissue,” explains Nantia Leonidou, first author of the current study.

Observing host metabolism after viral infection in the model
Using their integrated systems biology model to simulate infection with SARS-CoV-2 in bronchial epithelial cells, the researchers can identify host-based metabolic pathways that can be inhibited to suppress viral replication. “If you know the composition of a virus, you can run different scenarios and see how the biochemical reactions in host cells change during viral infection,” Dräger says. The team developed high-quality software to simulate an infection in a cell-type-specific manner.

New targets identified
Applying the model to another cell type, the research group confirmed the previously identified target, guanylate kinase 1, and discovered other new biochemical targets with remarkable antiviral effects. The most promising new hit was CTP synthase 1. Inhibition of this enzyme in the model reduced viral growth by 62 per cent without affecting the maintenance of the human host cells. Both target molecules are closely linked to the structure of genetic material, which requires the same building blocks in both the virus and the host cells. Andreas Dräger’s team believes these findings provide a crucial basis for accelerating the development of viral inhibitors.

“Our models could represent a paradigm shift in drug development and accelerate the preclinical phase,” emphasises Nantia Leonidou, adding, “The methods are fully transferable to any virus and host cell type and are also commercially viable.”

Dräger’s group now plans to apply their methods to other viruses. The first inhibitors for their discovered enzymes will be tested in animal models for safety, toxicity, and efficacy.

Publication: Leonidou N, Renz A, Mostolizadeh R and Dräger A: New Workflow Predicts Drug Targets Against SARS-CoV-2 via Metabolic Changes in Infected Cells. PLOS Computational Biology 2023. doi: 10.1371/journal.pcbi.1010903

3rd place in the Youth Research Competition

Fri, 03 Mar 2023 16:00:00 +0100

We congratulate the two students Henrik Schick and Simon Straub for their success at the regional competition in North Black Forrest to Youth research 2023 in Nagold! With their modeling and computer simulation of the Spanish flu they convinced the judges and reached the third place in this year's competition.

The competition happened on March 2 and 3, 2023, in Nagold’s Stadthalle. From morning to noon on March 2, the 72 students presented their 41 projects to the judges - including five mathematics and computer science projects. From the afternoon until the evening, there were elective leisure activities. In the evening, the regional winners were announced. Afterwards, a dinner was held with the participants, jurors and people from the public. On March 3, the public is invited. Everyone could come to the hall and learn about the participants’ projects. The event ended with the award ceremony in the afternoon.

Further information:

Reconstruction of a genome-scale metabolic network of Staphylococcus lugdunensis

Tue, 20 Sep 2022 11:00:00 +0200

Staphylococcus lugdunensis is a Gram-positive and coagulase-negative human commensal. The bacterium can produce a polypeptide antibiotic called lugdunin. Lugdunin is active against several pathogens, particularly Staphylococcus aureus, known to cause severe human infections. Nasal colonization by S. lugdunensis has significantly reduced S. aureus rates. Therefore, the use of S. lugdunensis as a probiotic to combat S. aureus in affected patients is being investigated. However, this endeavor is complicated because S. lugdunensis is an opportunistic pathogen with high virulence. Although part of the normal skin flora, it has been found to cause infections similar to S. aureus. In her bachelor’s thesis, Ms. Selin Sahin created the genome-scale metabolic model, iSLN22SS, to better understand the metabolism of this unique and diverse bacterium. This model was created using various computer programs and scripts, optimized, and finally validated by experimental data.

First genome-scale reconstruction of Klebsiella pneumoniae HS11286

Mon, 22 Aug 2022 12:00:00 +0200

Klebsiella pneumoniae is a Gram-negative, facultatively anaerobic bacterium from the Enterobacteriaceae family. It grows on various human mucous membranes, such as the nose and intestine. Klebsiella pneumoniae belongs to the so-called ESKAPE pathogens (Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter spp.). The increasing development of antibiotic resistance among this group of pathogens poses a threat to individuals with compromised immune systems and healthy individuals. Thus, the reconstruction of a genome-scale model for Klebsiella penumoniae HS11286 was motivated by gaining a deeper understanding of the pathogen’s metabolism and building a helpful research tool that can aid in the development of new treatments. The model was validated in several ways, firstly by simulating different growth rates on experimentally determined media compositions, such as the synthetic nasal medium SNM3, and secondly by predicting essential genes.

Genome-scale metabolic model of Acinetobacter baumannii

Mon, 22 Aug 2022 11:30:00 +0200

Acinetobacter baumannii is a Gram-negative, strictly aerobic bacterium and can also be an opportunistic pathogen in humans. Since its discovery, it has caused severe infections, notably in health care facilities. It also has garnered tremendous attention because of its interaction with SARS-CoV-2, which produces drastic consequences in intensive care units. Because of the increasing antibiotic resistance and high mortality of A. baumannii, it has been classified as an urgent threat by the Centers for Disease Control and Prevention since 2019. Yufan Xia created a highly standardized and well-annotated genome-scale metabolic model (GEM) of A. baumannii ATCC 17978 in her bachelor thesis. This model aims to help to identify which genes are required for its virulence factors. She analyzed the model and verified it against presently available experimental data of existent gene tests and sole carbon and nitrogen source tests. With this model, its interaction with other bacteria and its response to drugs can efficiently be simulated in silico. The model also gives insights into various metabolic processes in A. baumannii ATCC 17978.

First metabolic model of Proteus vulgaris FDAARGOS_1507 (DSM 46228)

Tue, 14 Jun 2022 14:00:00 +0200

P. vulgaris is a Gram-negative, facultatively anaerobic bacterium. It was first discovered in 1885 due to its swarming ability and involvement in putrefaction. Since then, it has been discovered that P. vulgaris is a commensal of various different microbiomes throughout the human and animal bodies. The bacterium is classified as an opportunistic pathogen and can lead to many infections.

To further understand the metabolic processes and answer the question of what metabolites are essential for the bacterium to grow, Manuel O. Harke reconstructed the first metabolic model of P. vulgaris during his master’s thesis. The reconstruction and evaluation were done using various software tools and self-written scripts.

The model simulates growth on various media, such as the synthetic nasal medium SNM3 and an artificial urine medium. These media were minimized to identify the essential metabolites, and their respective components were recombined. This process resulted in seven different minimal component compositions.

The model was uploaded to the BioModels database and will not yet released upon publication under this accession: www.ebi.ac.uk/biomodels/MODEL2205300001.

First disputation of our group successfully completed

Fri, 13 May 2022 11:30:00 +0200

We are proud to announce that Alina Renz has successfully defended her doctoral thesis. She joined us as an undergraduate student, earned her master’s degree with us, and is our first doctoral student. Her work was embedded in CMFI and DZIF in the fight against bacterial and viral infections, including SARSCoV2.

To name just a few of her accomplishments: twelve publications in just three years, international coverage of our research on radio, television and newspapers. Congratulations to Dr. des. Alina Renz, our first Ph.D. student and great inspiration for the group!

New GEM of Finegoldia magna ATCC 29328

Tue, 15 Mar 2022 11:45:00 +0100

Finegoldia magna, a Gram-positive, anaerobic coccus (GPAC), is a bacterium of clinical interest due to multiple possible infection sites in the human body and its variety of virulence factors. Moreover, it has belongs to the most pathogenic species among the GPACs. The bacterium is often the cause of dangerous infections because it acts as a commensal bacterium on the human skin and mucous membranes. In the hope of expanding the comprehension of F. magna, particularly in the nasal environment, a genome-scale metabolic model (GEM) was drafted, curated, and evaluated by Josua Carl in his bachelor's thesis. To this end, Josua Carl applied various software tools and self-implemented methods, which he made available on github.com/JosuaCarl/Script_Assisted_Modeling. He used the resulting model to simulate growth on the synthetic nasal medium three (SNM3) and in the minimal media to understand the required environment better.

New genome-scale reconstruction of Corynebacterium simulans PES1

Tue, 16 Nov 2021 12:45:00 +0100

Corynebacterium simulans is a Gram-positive aerobic bacterium belonging to the Actinobacteria group. It was first described in 2000. C. simulans is part of the typical human bacterial flora and can be found, for example, in the nose. Due to its close relationship to the pathogenic bacterium Corynebacterium striatum, it is sometimes misclassified as Corynebacterium striatum in the case of infection and can cause an incorrect therapeutic approach. C. simulans can act opportunely as a pathogen itself, but the mechanisms and conditions for this are largely unknown. For this reason, Jan-Philipp Leusch developed and reconstructed a genome-scale model of C. simulans PES1 in his bachelor thesis. Such a computer model can be used in the future to determine and further investigate the role of C. simulans and the interaction with other organisms in infections.

ILW Award 2020/21 goes to Nantia Leonidou for her master thesis

Tue, 26 Oct 2021 11:15:00 +0200

The joint board of Informatics in the Life Sciences of the German Informatics Society e. V. (GI) and the German Society for Medical Informatics, Biometry and Epidemiology e. V. (GMDS) honors Nantia Leonidou for her master thesis on tissue-specific reconstruction of restriction-based metabolic models.

Congratulations to Nantia Leonidou for receiving an award for her master’s thesis on “Tissue-specific reconstruction of constraint-based metabolic models based on ReconX” in the field of Bioinformatics and Medical Informatics at the 66th Annual Meeting of the GI and GMDS.

Successful Ph.D. defense in computational neuroscience

Wed, 15 Sep 2021 13:00:00 +0200

We congratulate our team member, Martina Feierabend, for successfully defending her Ph.D. thesis “Plasticity of the auditory modality” today at Tübingen’s Hertie-Institute of the TueNeuroCampus in Clinical Neuroscience.

Co-organization of the 6th annual SysMod meeting of the ISMB/ECCB.

Fri, 30 Jul 2021 18:00:00 +0200

The annual ISMB conference (“Intelligent Machines for Molecular Biology”) retook place in 2021 under corona conditions as a virtual event in combination with the European Bioinformatics Conference (ECCB). Since the foundation of the associated Symposium on Systems Modeling (SysMod) in 2016, the Tübingen working group on systems biology has contributed significantly to the organization and design of this event.

Members of the working group gave two talks:

Dr. Reihaneh Mostolizadeh: “Workflow for modeling microbial community interactions applied to Dolosigranulum pigrum and Staphylococcus aureus within the human nose.”
Nantia Leonidou: “Tissue-specific reconstruction of constraint-based metabolic models based on ReconX.”

Moreover, the working group was involved in the preparation of two others:

Dawson D. Payne: “An updated genome-scale metabolic network reconstruction of Pseudomonas aeruginosa PA14 to characterize mucin-driven shifts in bacterial metabolism.”
Sanjeev Dahal: “Genome-scale modeling of Pseudomonas aeruginosa PA14 unveils its broad metabolic capabilities and role of metabolism in virulence and drug potentiation.”

Jun.-Prof. Dr. Andreas Dräger chaired the session on “Infectious disease modeling.”

In addition, the Tübingen working group presented further results in the form of three scientific posters:

Martina Feierabend: “High-quality genome-scale reconstruction of Corynebacterium glutamicum ATCC 13032.”
Alina Renz: “Curating and Comparing 114 Strain-Specific Genome-Scale Metabolic Models of Staphylococcus aureus.”
Jun.-Prof. Dr. Andreas Dräger: “The Systems Biology Simulation Core Library.”

The next event on SysMod is planned to take place in 2022 again as a conference with face-to-face lectures if conditions allow.

Using computer models to identify weak spots in a virus

Mon, 05 Jul 2021 10:44:35 +0200

Andreas Dräger supports science communication - here with in the Podcast Nachgefragt! beim Bioinformatiker of the Apotheken Umschau (a German free bimonthly health magazine, available in German drugstores). Here is the Podcast link (German only).

Automating the refinement of C. striatum

Tue, 18 May 2021 12:00:00 +0200

Over the last years, more and more multi-drug resistant strains of Corynebacterium striatum have been found in clinical settings. Due to antibiotic resistance and other mechanisms, these bacteria can lead to diseases like pneumonia and chronic obstructive pulmonary disease. Thus, any hints on how these bacteria work can help prevent diseases, especially in immunocompromised patients. Based on a previous draft model created by Tanja Urz, Famke Bäuerle worked on the further refined genome-scale metabolic model of C. striatum. With growth simulations on different media, Ms. Bäuerle investigated the behavior of C. striatum in different settings.

The refinement of genome-scale metabolic models requires a lot of manual curation work. Some of those tasks can, however, be abbreviated by complementary scripts or even be automated. In the second part of this research project, Famke Bäuerle worked on automating the refinement. A collection of python scripts was developed, which can be used and extended by all group members. These scripts cover fundamental analysis, such as numbers that always go into publications and growth simulations of media definitions collected by the group members. All scripts were collected in a Python module called refinegems and saved to a GitHub repository.

Protein sequence clustering with DIAMOND

Wed, 03 Mar 2021 10:45:00 +0100

There has been an exponential increase in protein sequences in the last decades due to large-scale projects aiming at sequencing unknown species. The number of protein sequences will continue to increase in the future due to large-scale projects such as the Earth BioGenome Project. Protein sequence clustering plays an essential role in analyzing this large amount of data efficiently. This method enables the reduction of large protein datasets and the identification of functional and evolutionary similarities between different proteins.

The graph-based Greedy Vertex Cover algorithm used for clustering was extended by the option of cascaded clustering in the course of the bachelor thesis. Moreover, the resource consumption was limited by storing the node graph externally and made individually adaptable to the user’s working memory.

The Greedy Vertex Cover algorithm’s scalability was tested on the NR database’s random samples with the newly added options. The results of DIAMOND were compared to the MMseqs2 tool, which is currently the fastest and best tool for clustering large datasets. It was shown that DIAMOND is faster than MMseqs2 when parameters are kept comparable.

High school students model multidrug-resistant bacteria for the Youth Research Competition

Fri, 26 Feb 2021 20:00:00 +0100

We congratulate Aleyna Murat and Burcu Karakum on winning the 3rd award and two special prizes in the Youth Research Regional Competition North Black Forest to work on multi-resistant bacteria. Mentored by Alina Renz, the Otto-Hahn-Gymnasium students in Nagold investigated both Acinetobacter baumannii and Enterococcus faecium using genome-scale computer models systems biology methods. With their approach, they could suggest targets for potential treatment options against these two multi-resistant germs.

Tissue-specific reconstruction of constraint-based metabolic models based on ReconX

Tue, 26 Jan 2021 12:00:00 +0100

In a study published in October 2007, scientists studying coronaviruses characterized China’s situation as a ticking “time bomb” for a potential virus outbreak. They had three strong indications to worry: the animal-related eating habits in southern China, the previous appearance of SARS-CoV-like viruses in horseshoe bats, and the ability of coronaviruses to undergo recombination. Eighteen years later, the whole world experiences the realization of this prophecy with the emergence of COVID-19, one of the deadliest respiratory diseases. In this regard, scientists globally try to understand the host’s immunopathological response, how the virus adapted to new hosts, and how it spreads. Currently, great efforts are made to detect effective antiviral treatments for coronaviruses. Identifying potential antiviral targets is of great interest. One way to detect them is by analyzing metabolic changes in infected cells. In 2012, Wang et al. published mCADRE (metabolic Context-specificity Assessed by Deterministic Reaction Evaluation), aiming to reconstruct tissue-specific models using gene expression data and network topology information. The algorithm is implemented in MATLAB, and its functionality is based solely on the first version of the human model. This resulted in its limited usability in the last few years since newer and more comprehensive model versions are available.

In her master thesis, Nantia Leonidou created pymCADRE, a re-implementation of mCADRE in Python 3.8, and tested its functionality using all three currently available versions of the human metabolic network. From this, it was observed that internal optimizations done with fastFVA resulted in context-specific models closer to the ground truth.

Furthermore, viruses consume energy from the host cell to increase their mass production. Hence, they genetically re-program cells to form further virus particles and enable their reproduction. Understanding the metabolic basis of host-virus interaction could be used to predict the metabolic changes and their impacts on virus reproduction. For this reason, host-virus models (HVMs) were created with pymCADRE to help to identify potential antiviral targets against SARS-CoV-2. With those models, the existence of the recently identified potential antiviral target enzyme guanylate kinase was further verified.

Bioinformatics-based potential therapeutic approach against COVID-19

Thu, 14 Jan 2021 10:44:00 +0100

Proliferation of the COVID-19 virus is only possible through reprogramming the metabolism of its human host and exploiting their enzymes. With the help of a detailed computer model by the Computational Systems Biology of Infection group of Jun.-Prof. Andreas Dräger it is now possible to identify the human enzyme which is ncessary for COVID-19 virus proliferation, but whose function is not essential for cell preservation. Thus this enzyme may be a suitable target for an antiviral agent. The results of this study have been published in the current issue of the Bioinformatics journal (FBA reveals guanylate kinase as a potential target for antiviral therapies against SARS-CoV-2).

In cooperation with the Fraunhofer Institute for Molecular Biology and Applied Ecology IME in Hamburg, scientists are now investigating whether medical treatment targeting this enzyme will inhibit coronavirus proliferation.

Official press release by the German Center for Infection Research (DZIF):
SARS-CoV-2: Bioinformatiker entdecken eine neue Schwachstelle des Virus (German only)

Interview with the Tübingen University:
Potentielles Wirkstoffziel: Wie das Ausschalten eines Enzyms das Virus SARS-CoV-2 stoppen könnte (German only)

Here a selection of media features:
- n-tv: Durchbruch ebnet Weg zu Corona-Medikament (German only)
- SWR: Neuer Ansatz für Corona-Therapie an Tübinger Uni (German only)
- Berliner Zeitung: Covid-19: Bioinformatiker entdecken Schwachstelle des Coronavirus (German only)
- MDR: Jena, Dresden, Tübingen - vielversprechende Ansätze für ein Corona-Medikament (German only)
- Computer Bild: Coronavirus: Deutsche Bioinformatiker entwickeln Therapie (German only)
- bTV: Bulgarian TV feature (Bulgarian only)
- Schwäbisches Tagblatt: Durchbruch im Kamp gegen das Virus? (German only)
- c’t 2021, issue 3, p 40: https://www.heise.de/select/ct/2021/3 (German only, registration required)
- Isna: Feature in the Iranian Students' News Agency (Persian only)

Genome-scale model of Corynebacterium glutamicum

Wed, 21 Oct 2020 11:00:00 +0200

The bacterium Corynebacterium glutamicum can produce ʟ-glutamic acid under certain growth conditions. ʟ-Glutamic acid is used as a flavor enhancer, food supplement, or primary chemical raw material. It, therefore, plays an essential economic role with an annual production of over 2½ million tons. Due to metabolic development, the product range of C. glutamicum has been expanded to include all biogenic amino acids, vitamins, and more. Metabolic models of C. glutamicum already exist. These are to be supplemented with new metabolic data and expanded with data from new systems biology programs.

For this purpose, Martina Feierabend created general bacterial metabolism model, which was gradually supplemented by metabolites and reactions typical for this organism in her master thesis in collaboration with experts from the Research Center Jülich. The SBML document was supplemented and extended by MIRIAM standards (MIRIAM: Minimum Information Required in the Annotation of Models) and continuously improved by memote. The computer model was examined on different growth media and tested on ʟ-glutamic acid production. The simulation results show that C. glutamicum already has bacterial properties, but these are typical for the species.

Automating the Assignment of Systems Biology Onotology Terms

Wed, 07 Oct 2020 13:00:00 +0200

Annotations, such as the Systems Biology Ontology (SBO) terms, play an essential role in constructing computational models in SBML format. The thesis of Elisabeth Fritze aimed to automate SBO term annotation for reactions in SBML models. To this end, an annotation program was implemented in Python. In the process, Ms. Fritze also formulated a mapping between Enzyme Commission (EC) numbers to SBO terms to enable SBO terms to EC annotated reactions.

Additionally, the annotation script was tested on the 108 available models from the so-called BiGG database. This test showcased the gained diversity and specificity in SBO terms across the models. Thus, her bachelor thesis results provide additional context to models and can be seen as the first step for automated reaction analysis based on SBO terms.

Visual exploratory web application for flux sampling results

Tue, 15 Sep 2020 12:15:00 +0200

Constraint-based modeling methods can predict the metabolic phenotype of cells and organisms on a molecular level. However, in several unstable environments, a cellular goal cannot be easily identified. Consequently, flux sampling is an unbiased constraint-based modeling method applied in such situations. It returns information about reaction flux ranges, and their probability of a genome scaled metabolic model (GEM). Flux sampling simulations generate a lot of data from GEMs. Hence, data-driven methods are needed to analyze results as a whole, instead of focusing on a subset of known essential reactions.

In his master thesis, Constantin Holzapfel designed and implemented a solution for a data-driven exploratory web application. The resulting software can visualize and analyze flux sampling simulation results using current data visualization principles directly within common web browsers. A combination of standardized scoring of flux differences, flux distributions, and pathway network visualizations was used to guide the analysis process according to Shneiderman’s visual information-seeking mantra “Overview first, zoom in and filter, details on demand.”

In a case study, the red blood cell (RBC) metabolism was studied using the GEM “RBC iAB_RBC_283” in varying blood plasma environments. Constantin Holzapfel constrained this GEM with extracellular metabolomic measurements. To this end, athletes’ blood samples were taken at the following time points: before and after high-intensity training and two hours after exercise.

The simulation results showed the impact of glucose at these time points and how reduced glutathione levels change to provide higher protection against reactive oxygen species during training.

Alina Renz awarded for her master thesis about virulence factors of P. aeruginosa

Wed, 09 Sep 2020 15:45:00 +0200

We congratulate Alina Renz for the recognition of her master thesis “Modeling of Potentially Virulence-Associated Metabolic Pathways in Pseudomonas aeruginosa PA14 including experimental verification” with the Medical Bioinformatics Award of the German Society for Medical Informatics, Biometry and Epidemiology (GMDS e.V.) in the category Medical Bioinformatics.

Diagramming Against COVID-19

Wed, 01 Jul 2020 21:04:35 +0200

Supported by the local company yWorks in Tübingen and with a great international team of collaborators, we contribute to the global fight against COVID-19. To this end, we work on building a knowledge representation of the known molecular mechanisms at all stages of the development of the virus and its interaction with its host. For more information, visit https://www.yworks.com/support/fightcorona and https://twitter.com/CovidPathways.

Genome-scale Model of Staphylococcus epidermidis

Tue, 16 Jun 2020 14:30:00 +0200

The human body offers niches for numerous microorganisms, many of which can promote health or cause diseases. One frequently isolated member of the skin microbiome, Staphylococcus epidermidis, can play both roles and may cause severe inflammation. How exactly the bacteriumʼs metabolic response to environmental conditions changes was the question, to which Anastasiia Grekova dedicated her bachelor thesis.

She developed a metabolic model of the bacterium. The laboratory data obtained from the literature enabled her to analyze the growth behavior of the model. In particular, the model appeared to be oxygen sensitive and to activate different metabolic pathways depending on the conditions. The computer model was examined on different nutrient media, such as a nose-like synthetic medium. Further insight into the ecological adaptation of S. epidermidis was gained using a collaborative model with Staphylococcus aureus. The simulation results showed that the bacteria have to compete for some amino acids that are deficient in the nasal medium. Thus, interaction reduces the growth rate.

The new model design of S. epidermidis expands the ensemble of the in silico nasal microbiome, which provides a fundamental understanding of antimicrobial-resistant pathogens.

First draft model for Corynebacterium striatum KC-Na-01

Tue, 02 Jun 2020 14:00:00 +0200

The Gram-positive bacterium Corynebacterium striatum can lead to opportunistic infections, particularly in immunodeficient patients. The increasing mortality associated with C. striatum is due, among other things, to the growing antibiotic resistance of the bacterium. For reasons that remain to be fully clarified, C. striatum can reduce the virulence of the pathogen Staphylococcus aureus, which is also found in the nose, when colonized nasally, although it has pathogenic properties. Thus, C. striatum could be a suitable prebiotic, as it could positively influence the behavior of the much more dangerous bacterium S. aureus towards commensalism.

In this research project, Tanja Urz presented a first model for C. striatum. The project aimed to reconstruct a genome-scale metabolic model (GEM) of C. striatum KC-Na-01, which will help better to understand the mechanisms of interaction with nasal S. aureus. Furthermore, nasal microbiomeʼs behavior can be simulated to identify possible competitive mechanisms that may serve as clinical therapy approaches.

Visualization of clinical multiomics data—The web-based application VMOD is extended

Tue, 02 Jun 2020 13:30:00 +0200

Visualization of Multiomics Data, VMOD for short, is a visual web-based analysis application. Based on a transcriptomics data set, it enables the exploration of different networks of biological pathways. All data is processed locally, without the need to install additional software.

Based on Manuel Harkeʼs bachelor thesis (2019), Ebru Cobanoglu added further functions in her bachelor thesis. VMOD improved by the possibility of exploring single metabolic pathways with an arrangement known from KEGG. Furthermore, additional information from the KEGG database is graphically displayed and linked.

Tabular editor for systems biology models

Wed, 20 May 2020 23:00:02 +0200

During her research project, Marietta Hamberger further developed and redesigned the SBMLSheets plugin initiated by Robert Deibel for the InSilico project.

The file format SBML, which is frequently used in systems biology, is based on XML. It is, therefore, not intuitively accessible to all users, the SBTab project follows the approach representing the information contained in SBML in tabular form. SBTab files adhere to specified table definitions. Therefore, it is still necessary to read the SBTab documentation to create correctly formatted documents.

A user interface has been developed to simplify this process for creating, modifying, and converting SBTab documents. The table editor supports the import and export of data (CSV, TSV, and TAB formatted), the conversion of tables into SBTab and SBML documents, and other useful functions. So far, three table types (Compound, Compartment, and Reactions) are covered. In the future, however, SBMLSheets will be completed to ensure all table types and a complete SBML conversion.

Automatic linking from genome to the reactome

Tue, 05 May 2020 13:30:00 +0200

Manuel Dienert successfully defended his bachelor's thesis. Therein, he dealt with the question of how gene-protein reaction rules (GPRs) for systems biological models of any organism can be automatically derived. GPRs link genes with reactions and thus enable metabolic analyses on the genome level. This includes, for example, the prediction of possible phenotypes after gene knockouts.
The developed approach automatically derives GPRs for a part of the reactions of tested models. For this purpose, Mr. Dienert uses annotated genomes and links the genes to reactions of the model using EC numbers with which reactions and genes are often annotated. The method was implemented in Java™️ using the library JSBML.
The results provided by his work provide a valuable starting point for model reconstructions in future projects.

Google funds 2 students' projects

Mon, 04 May 2020 20:00:00 +0200

Google is sponsoring two international students to participate in software projects of the Systems Biology group within the Summer of Code 2020: Hemil Panchiwala will extend the simulation software SBSCL and add functions for stochastic simulations, among other things. Hemand Yadav will take care of lossless file conversion from SBML to the also frequently used JSON format.

First model draft of Dolosigranulum pigrum 83VPs KB5

Mon, 16 Mar 2020 12:00:00 +0100

Dolosigranulum pigrum is a Gram-positive bacterium. As pathobiont, the bacterium is mainly a commensal of the upper respiratory tract, but may also be isolated from various sites of infection in immunodeficient patients. What makes this organism so interesting is that it is a negative predictor for Staphylococcus aureus colonization of the nose. Although the mechanism behind the inhibition has not been elucidated, D. pigrum may be a suitable candidate for the exploration as probiotic. Additionally, the uncovering of the mechanism behind the inhibition may bring about a new antibiotic or competitive mechanism exploitable in the clinic. For this promising potential, Lina Widerspick constructed a genome-scale reconstruction (GEM) of D. pigrum 83VPs KB5 within her research project. Such a computer model may serve as a knowledge-base of the organism and can be employed as a well-curated foundation for a simulation of the colonized nose.

Web-based visualization of clinical multi-omic data in networks

Tue, 19 Nov 2019 11:00:00 +0100

With the first version of VMOD (Visualization of Multiomics Data), Manuel O. Harke completes his bachelor thesis. The web-based software is suitable for clinical use because all data is processed locally. Because modern web browsers with JavaScript interpreters are integrated as standard in current operating systems, it is not necessary to install additional software. The first version of the software, which is implemented in a server/client architecture, first loads gene expression data and displays them in the context of known KEGG maps. The modular implementation allows an extension to further data sets, implementation of export functions for interoperability with similar programs as well as improved layout functions.

Antibiotic resistance in Pseudomonas aeruginosa

Tue, 24 Sep 2019 10:45:00 +0200

Pseudomonas aeruginosa is a Gram-negative, opportunistic pathogen. As such, it is responsible for infections in hospitalized patients with local or systemic immunodeficiency. A notable problem of infections with P. aeruginosa is the high rate of antibiotic resistance. A fundamental understanding of these resistance mechanisms is required to produce new, effective antibiotics against this pathogen. In his bachelor thesis, Eike Pertuch investigated modifications of lipopolysaccharides (LPS), which are components of the outer bacterial membrane and whose modifications can reduce the efficacy of antibiotics. He extended an existing genome-scale model of P. aeruginosa to include the different LPS modifications and then analyzed the altered metabolism for different modifications. His extensions and initial analyses expanded and deepened the understanding of the resistance mechanisms in P. aeruginosa and can pave the way for new, innovative treatment strategies in further analyses.

First model draft of Moraxella catarrhalis BBH18

Tue, 10 Sep 2019 11:45:00 +0200

The aerobic Gram-negative bacterium M. catarrhalis causes a large number of respiratory tract infections and was discovered as early as 1896. The spread of antibiotic resistance makes treatment increasingly difficult. Moreover, it has not yet been possible to develop an effective vaccine against this pathogen. New promising therapeutic approaches could be identified by computer simulation. For this purpose, a genome-scale metabolic model for M. catarrhalis is required. As part of her research project, Pia Rautenstrauch succeeded for the first time in designing such a model specifically for M. catarrhalis strain BBH18. This represents the first step towards a comprehensive model of the human nasal microbiome.

Updated model of the pathogen H. influenzae

Tue, 10 Sep 2019 10:45:00 +0200

As a result of his bachelor thesis, Niklas Henle updated the model of the pathogen Haemophilus influenzae, 20 years after Schilling et al. created the original model. The updated model now contains significantly more metabolic pathways and is available in the current version of the systems biology file format SBML.

Moreover, Mr. Henle thus fine-tuned the new version of the model to predict cell growth on the synthetic nasal fluid SNM3, which Krismer et al. experimentally defined in Tübingen. This medium provides all requirements of growth for a variety of different nasal bacteria. Mr. Henle's research can pave the way toward a comprehensive model of the nasal microbiome.

Google announces results of the Summers of Code 2019

Tue, 03 Sep 2019 22:45:00 +0200

The three Indian students, Nikhil Ghodke, Kaustubh Trivedi, and Bhavye Jain, who worked externally with the Systems Biology Working Group of the University of Tübingen over the summer, achieved the goals of this year's Google Summer of Code, as has just been announced.

This project enabled the development of new versions of existing software as well as the basis for the development of new software. For all participants, the project was not only a significant step forward in software development but also marked by cultural exchange. In addition to the former students Roman Schulte and Thomas J. Zajac as well as Thomas M. Hamm and JProf. Dr. Andreas Dräger from the Tübingen Systems Biology working group, mentors from other international working groups also participated in the projects: Dr. Matthias König (Humboldt University), Nicolas Rodriguez (Babraham Institute) and Dr. Lucian P. Smith (University of Washington, Seattle).

The following software of the working group was contributed:

In the Java library JSBML a module was created for better support of spatially structured biomodels.
The biomodel software ModelPolisher has been updated and is much more user-friendly.
The foundations for increased developer-friendliness of the plugin framework InSilicoLab were laid.

Dynamically animated metabolic networks

Thu, 29 Aug 2019 00:35:00 +0200

With version 2.0, the SBMLsimulator software supports not only the latest SBML standard, including numerous extension packages but also the graphical representation of metabolic networks in the form of process diagrams of the current SBGN specifikation. For the first time, time-series data can be processed into interactive animations and saved as video sequences. Since the software supports many widespread formats, the results are ideal for publication, analysis, or integration into presentation slides. More details can be found in this article.

Interactive visual simulator for systems biology models

Mon, 17 Jun 2019 15:00:00 +0200

As a result of his bachelor thesis, Mykola Zakharchuk created a new plugin for the InSilico platform with which systems biology models can be simulated with a mouse click and the results can be displayed graphically. The software fits easily into the InSilico framework and supports both SBML and SED-ML as well as files as COMBINE archives (OMEX). Thus even complex models can be simulated intuitively. The graphical representation of the results supports different axis scaling (logarithmic or linear) as well as export to standard image formats. The software is freely available on github.com/draeger-lab/vissim/.

Google funds three students' projects

Mon, 06 May 2019 21:15:00 +0200

Google supports three international students to contribute to software projects of the systems biology research group:

Nikhil Ghodke extends the plugin framework InSilico.
Kaustubh Trivedi improves the annotation tool ModelPolisher.
Bhavye Jain works on the library JSBML for the validation of spatial models.

4th Annual Meeting for Biological Systems Modeling

Mon, 11 Mar 2019 20:15:00 +0100

We invite you to participate in the 4th Annual SysMod meeting that bridges systems biology and bioinformatics at ISMB/ECCB on July 22 in Basel, Switzerland! Submit your abstract by April 11 and share your work on the development of models of cells, tissue, and organisms with the community. For more details, visit sysmod.info.

First version of an InSilico app for SBML files

Sat, 02 Mar 2019 10:34:14 +0100

The SBTab table format developed for Excel is very well suited for processing systems biology models. In his bachelor thesis, Robert Deibel designed and created a first version of an app for the InSilico framework, with which SBML files can be mapped to tables and thus edited.

Google Summer of Code 2019

Tue, 26 Feb 2019 22:03:35 +0100

Also this year Google accepted our application. Under the umbrella organization NRNB (National Resource for Network Biology), international students can now apply for advertised projects. Success is achieved by contacting potential mentors at an early stage. We wish you much success and fun with your programming!

Further information

Gene deletion increases bacterial virulence

Wed, 19 Dec 2018 10:15:00 +0100

In her master thesis, Alina Renz extended the recently published model of the widespread pathogen Pseudomonas aeruginosa, which often appears as a multi-drug resistant hospital germ. Using systematic computer simulation, she detected promising candidate genes whose absence was supposed to increase the virulence of the pathogen. In collaboration with the research group Schütz/Bohn at the Interfaculty Institute for Microbiology and Infection Biology independently conducted wet-lab experiments, evidence for these predictions could be collected. This thesis made a significant contribution to the cooperation between computer simulation and experimental work.

The first draft model of the syphilis pathogen

Tue, 30 Oct 2018 11:00:00 +0100

In cooperation with the University of Milan BICOCCA and the chair of Prof. Nieselt , the first draft for a model of the central carbon change of the syphilis pathogen Treponema pallidum ssp. pallidum was developed. Ms. Morini presented the results of the work to a professional audience of modelers during the COMBINE-Conference 2018 in Boston.

New cluster of excellence for fighting infectious diseases

Thu, 27 Sep 2018 16:15:00 +0200

The German Research Foundation (DFG) announces the 57 clusters of excellence in Germany, which will receive a total of 2.7 billion euros in funding for seven years from January 2019. The University of Tübingen accounts for three of these clusters. Our research group is part of the cluster for fighting infectious diseases, and our Institute of Computer Science will receive a cluster for machine learning. The application process took two years. The University of Tübingen has a good chance of remaining one of Germany's eleven distinguished universities with the title of excellence. Further information can be found in the press release of the University of Tübingen.

Visualization of Biochemical Processes - Temperature Dependence in Red Blood Cells

Mon, 24 Sep 2018 17:30:00 +0200

This time we look at the temperature dependence of the metabolism in human erythrocytes. This short animated video highlights metabolic effects in blood. The video explains all processes easily understandable and step by step in 1½ minutes.

All shown processes are based on real measured values, which were evaluated in a cooperation between the University of Tübingen, the University of California at San Diego and the Icelandic Blood Bank in the context of the Bachelor thesis of Lea Buchweitz.

Visualization of biochemical processes—storage of blood platelets

Thu, 20 Sep 2018 17:45:00 +0200

A computer animation shows the biochemical processes that take place in blood platelets when they are stored in blood banks. For this purpose, a new visualization technique was developed that enables the entire cell metabolism of human thrombocytes to be observed at the same time. The video explains all processes easily understandable and step by step in less than 8½ minutes. All processes shown are based on real measurements that were evaluated in a cooperation between the University of Tübingen, the University of California at San Diego and the Icelandic Blood Bank as part of Lea Buchweitz's bachelor thesis.

New plugin-based software framework

Thu, 23 Aug 2018 15:00:00 +0200

In his thesis "InSilico - Concept of a Modular Research Environment," the computer science student Roman Schulte developed a new software infrastructure based on the Eclipse project and JavaFX under the co-supervision of Dr. Matthias König from Humboldt University Berlin. The new framework enables the creation of unified graphical user interfaces while at the same time reducing the effort required for software development. In the long run, less maintenance can be expected. The open source framework can be freely downloaded from our software page: https://github.com/draeger-lab/insilico.

Google Summer of Code 2018 (GSoC)

Wed, 22 Aug 2018 17:00:00 +0200

Supported by Google, the Java™ library SBSCL for systems biology simulations, which was developed in Tübingen since 2007, has been extensively updated this summer by the external Ph.D. student Shalin Shah from Duke University in Durham (North Carolina, United States). Thanks to international cooperation under the auspices of the NRNB (National Resource for Network Biology) with Dr. Matthias König from Humboldt University in Berlin and Nicolas Rodriguez from the Babraham Institute in Cambridge (United Kingdom), numerous innovations have been implemented. SBSCL is now entirely based on Maven and, thanks to the support of Dr. Hannes Planatscher, can now be used independently of commercial solution methods. Support for the SED-ML format has also been revised.

Newly appointed: Junior Professor Dr. Andreas Dräger

Mon, 02 Jul 2018 11:30:00 +0200

Dr. Andreas Dräger (born 1980) was appointed Junior Professor of Computational Systems Biology of Infection and Antimicrobial-Resistant Pathogens at the Faculty of Science in May 2018. Since 2016, he has led a junior group at the Chair of Applied Bioinformatics at the Center for Bioinformatics (ZBIT) at the University of Tübingen.

Andreas Dräger studied bioinformatics at the Martin Luther University in Halle. During his studies, he completed research internships at the Max Planck Institute for Molecular Genetics in Berlin, at the Leibniz Institute for Plant Biochemistry in Halle and the University of Illinois in Chicago (USA). His doctoral thesis on "Computational Modeling of Biochemical Networks," which he wrote at the Tübingen Chair of Cognitive Systems and the Keio University in Yokohama (Japan), was honored by the Faculty of Natural Sciences with the 2011 Ph.D. Award and a position as a junior research group leader at ZBIT. After two-year research stay at the University of California in San Diego (USA), he returned to the University of Tübingen. Here, he supplemented his teaching with lectures, tutorials, and internships in systems biology.

His research focuses on the question of how biological systems can be simulated in computers. For example, Andreas Dräger was involved in a project that, for the first time, shows all human cell metabolic processes three-dimensionally in computers. This enables a detailed understanding of drug intolerances or metabolic diseases, for example. He now wants to combine this human model with models of germs that are difficult to treat to understand the development of infections and discover new therapeutic approaches.

Further information

Modeling of metabolism and gene expression in SBML

Tue, 26 Jun 2018 12:15:00 +0200

In his master thesis, Marc Alexander Voigt laid the foundation for a uniform standard for the description of ME models in the SBML file format. These are systems biology models that link metabolism with effects of gene expression. For a long time, these processes could only be described separately. When models were finally developed that combined both, there was a lack of a suitable format for data exchange between different simulators. The work was carried out in close cooperation with colleagues from the University of California in San Diego and also in consultation with colleagues from Paris and Luxembourg.

Validation and simulation of qualitative models in Java™

Thu, 17 May 2018 14:15:00 +0200

In her bachelor thesis, Lisa Falk extended the widely used library JSBML by a package for the representation of logical models. Now Boolean models, Petri nets and much more can be processed. Lisa Falk implemented the necessary validation rules to ensure the syntactical correctness of the models. In cooperation with Prof. Claudine Chaouiya, it was also possible to automatically derive equations for function terms and integrate them into SBMLsqueezer. This means that logical models can now be created partially automatically. The work was done in close cooperation with Nicolas Rodriguez and builds on the results of Roman Schulte's project in the Google Summer of Code 2016.

Dynamic visualization of time-dependent data in the context of systems biology networks

Thu, 28 Sep 2017 14:15:00 +0200

Lea Buchweitz and Christoph Blessing developed two different methods for the visual representation of temporally resolved data in the context of biological networks in their bachelor theses. Models of perception, which were represented in forms and colors, were taken up. As a result, both the desktop program SBMLsimulator and the online tool Escher now support time series data. Further information can be found in the YouTube channel of the research group.

Modeling of the essential genes of microbial life

Thu, 31 Aug 2017 10:15:00 +0200

As part of the EU project AMBiCon, a study was published that characterizes the vital genetic nucleus of the bacterium Escherichia coli for the first time. The method can be transferred to other microorganisms and other cells. Thus, it is now possible to determine which cellular functions are essential for survival. Potential applications include medicine, food production, and biotechnology.

Further information

YouTube Channel to Systems Biology

Sun, 11 Sep 2016 11:32:00 +0200

Now there is a new YouTube channel with topics around systems biology. The offer is continuously being expanded.

https://youtube.com/c/systemsbiology

Tübingen Bioinformatics supports international study into microbe survival

Wed, 12 Aug 2015 15:56:00 +0200

University of Tübingen bioinformatics specialist Dr. Andreas Dräger is part of a team led by bioengineers at the University of California, San Diego, which has defined the core set of genes and functions that a bacterial cell needs to sustain life. The research, which answers the fundamental question of what minimum set of functions bacterial cells require to survive, could lead to new cell engineering approaches for E. coli and other microorganisms, the researchers said.

The findings are published online in the early edition of Proceedings of the National Academy of Sciences the week of August 10, 2015. This core set of genes is “the smallest common denominator that microbes need to have to become functional,” said Bernhard Palsson, the Galetti Professor of Bioengineering at UC San Diego and corresponding author on the paper. “If the cell lacks any of the genes from this set, the cell can neither function nor survive.”

According to the researchers, these findings could open up new avenues for cell engineering applications. Consider, for example, the genetic engineering of microbes to make value-added chemicals. This engineering process is typically done by making changes to the genetic makeup of a cell, which can end up toying with the cell’s core genes and functions, resulting in a “sick” cell.

Rather than risk compromising the cell’s core genes and functions, a new engineering approach could involve building the cell starting with the core set and adding on the extra desired functions, like chemical production. The PNAS paper presents the minimum core components that are absolutely necessary to include in the blueprints of an engineered cell.

“By defining the vital set of genes and functions that need to always be present in a cell to sustain life, we can begin to realize new ways to engineer a cell to optimize production of a desired product without sacrificing the cell’s health,” said Laurence Yang, a postdoctoral researcher in Palsson’s Systems Biology Research Group at UC San Diego and a co-first author of the paper.

The work, led by Palsson’s research group at UC San Diego Jacobs School of Engineering, is a collaborative effort with numerical and statistical experts from Stanford University. The researchers defined the core set of genes and functions as the “paleome,” referring to the ancestral genes and proteins that are at the heart of sustaining life for microbial cells.

“Other approaches have tried to define the paleome by comparing genome sequences and finding the gene portfolio that seemed to be similar in all of these sequences. This just defines the minimal genome. Our definition of the paleome takes a more comprehensive approach. It is a systems-biology-based definition that takes into account not just the minimum set of genes, but also the minimum set of functions, reactions and processes needed to build a cell,” said Palsson.

The team’s approach to define the paleome is based on a genome-scale computational model for cellular growth in E. coli. The researchers developed this model to account for all the metabolic and gene expression processes in the cell. Using this model, the researchers simulated the growth of a well-studied strain of E. coli across 333 different growth conditions. In each simulated growth condition, the main nutrient source of the growth medium (carbon, nitrogen, phosphorus, or sulfur source) was varied. The team observed which set of genes was consistently expressed throughout all the different growth environments and used this set to construct the paleome. In total, the team identified 356 genes that were expressed in all of these simulations.

“Our paleome definition is representative of core function not only in the well-studied strain of E. coli, but also in another strain of E. coli and three other microorganisms. We are hoping to use this paleome as a starter kit to rapidly build a new generation of genome-scale cellular growth models for other organisms,” said Yang.

“This study is an example of what’s called a ‘Big Data to Knowledge’ study,” added Palsson.
“We are demonstrating that we can take large data sets, integrate them together and analyze them to generate knowledge. In this case, we have used large amounts of experimental data and integrated them in the form of a computational model to arrive at our systems biology definition of the paleome.”

Publication:

Laurence Yang, Justin Tan, Edward J. O’Brien, Jonathan M. Monk, Donghyuk Kim, Howard J. Li, Pep Charusanti, Ali Ebrahim, Colton J. Lloyd, James T. Yurkovich, Bin Du, Alex Thomas, Andreas Dräger und Bernhard O. Palsson: Systems biology definition of the core proteome of metabolism and expression is consistent with high-throughput data. Proceedings of the National Academy of Sciences (PNAS), online publication, 10. August 2015, DOI: 10.1073/pnas.1501384112

Contact:

Dr. Andreas Dräger
University of Tübingen
Center for Bioinformatics Tübingen
Sand 1, 72076 Tübingen, Germany
andreas.draeger[at]uni-tuebingen.de

Eberhard Karls Universität Tübingen

Public Relations Department

Dr. Karl Guido Rijkhoek

Director

Janna Eberhardt

Research Reporter

Phone +49 7071 29-76753

Fax +49 7071 29-5566

janna.eberhardt[at]uni-tuebingen.de

www.uni-tuebingen.de/aktuelles

Bioengineers identify essential genes of microbial life.

Mon, 10 Aug 2015 11:45:00 +0200

In collaboration with the University of California at San Diego, a new study was published in the journal PNAS in which the minimal amount of genes and functions needed for a bacterial cell to survive was compiled for the first time.

These results could pave the way for much more effective genetic modification by preventing the cultivation of pathogenic microbes. Further applications are in medicine and food production.

Further information