Eldorado - Repository of the TU Dortmund
Resources for and from Research, Teaching and Studying
This is the institutional repository of the TU Dortmund. Ressources for Research, Study and Teaching are archived and made publicly available.
Communities in Eldorado
Select a community to browse its collections.
Recent Submissions
Improvement of protein quantification for proteins with shared peptides by using bipartite peptide-protein graphs
(2024) Schork, Karin Ulrike; Rahnenführer, Jörg; Eisenacher, Martin
In bottom-up proteomics, proteins are enzymatically digested to peptides, smaller amino acid chains, which are then measured via mass spectrometry. The received peptide quantities need to be summarized to protein quantities to obtain biological insights (protein quantification). This is complicated by the presence of shared peptides that occur in multiple protein sequences. The relationship between peptides and corresponding proteins can be represented as bipartite graphs. A novel protein quantification method, called bppgQuant, is proposed which calculates protein ratios from peptide ratios. It uses the structures of the bipartite peptide-protein graphs to build an equation system. As this system is not exactly solvable in many cases, an optimization problem is formulated to find solutions which minimize the sum of squared error terms. bppgQuant is evaluated and compared to the methods SCAMPI and PQP using four different quantitative datasets from different organisms, that contain known protein ratios. In summary, bppgQuant showed good results in comparison to SCAMPI and PQP, especially when protein nodes not needed to explain the measured peptide ratios were removed before optimization.
Coworking Best Practices for European Universities
(2025-02) Hensellek, Simon; Weißwange, Jonah; Orel, Marko; Válek, Lukáš; Diller, Sandra J.; Weber, Magdalena; Capdevila, Ignasi; Mérindol, Valérie
Verlässliche Identifikation logistischer Entitäten anhand inhärenter visueller Merkmale
(2025) Rutinowski, Jérôme; Kirchheim, Alice; Müller, Emmanuel
Im industriellen Umfeld und insbesondere in der Logistikbranche ist die Kenntnis über den Aufenthaltsort von Gütern und die Möglichkeit der Identifikation ebenjener von großem Nutzen. Sofern diese Informationen erhoben werden, geschieht dies durch die Nutzung extrinsischer Merkmale, wie etwa Barcodes. Jedoch sind häufig kostengünstige Entitäten, wie etwa Paletten, bislang nicht serialisiert und können somit nicht über ihren Lebenszyklus verfolgt werden. Die Serialisierung dieser Entitäten anhand ihrer inhärenten visuellen Merkmale bringt jedoch erhebliche Vorteile hinsichtlich der Nachverfolgbarkeit und Prozessoptimierung mit sich. Die dadurch gewonnene Prozesstransparenz wiederum stärkt das Vertrauen in sonst intransparente Vorgänge. Diese Promotionsschrift befasst sich deshalb mit der Bearbeitung der beiden folgenden Forschungsziele: Das erste Forschungsziel dieser Arbeit ist die Identifikation logistischer Entitäten anhand ihrer inhärenten visuellen Merkmale. Hierbei sollen geeignete logistische Entitäten ausgewählt und entsprechende Datensätze erstellt werden. Anschließend werden verschiedene Identifikationsmethoden auf die Datensätze angewendet und verglichen. Das zweite Forschungsziel ist die erstmalige Definition und anschließende Quantifikation des Begriffs der Verlässlichkeit im Maschinellen Lernen im Allgemeinen und des Identifikationsverfahrens im Spezifischen. Es wird eine Definition des Begriffs basierend auf seinen Komponenten, die aus der relevanten Literatur deduziert werden, entwickelt und eine geeignete Quantifikationsmetrik erarbeitet. Zuletzt werden die erstellten Datensätze qualitativ evaluiert und die Identifikationsmethoden anhand ihrer Prädiktionsgenauigkeit bemessen. Es wird weiterhin ein industrienahes Szenario der Identifikation zur Evaluation der Umsetzbarkeit des Verfahrens umgesetzt. Wiederum werden die Definition und Quantifikationsmetrik der Verlässlichkeit anhand ihrer Reliabilität in der Nutzung von Experten evaluiert.
Engineering a synthetic RNA segregation system
(2023-01-22) Hürtgen, Daniel; Mascarenhas, Judita; Vibhute, Mahesh A.; Weise, Laura I.; Mayr, Viktoria S.; Sourjik, Victor; Mutschler, Hannes
Cells possess a number of active segregation machineries for both chromosomal and large extrachromosomal DNA elements to avoid stochastic loss during cell division. In contrast, system that can be exploited for active, general segregation of RNA molecules including mRNAs or self-replicating RNA constructs are currently lacking. Here, we present an artificial RNA segregation system derived from the bacterial type II ParMRC plasmid segregation system and the RNA coliphage MS2. We show that fusing the partition protein ParR with the MS2 RNA coat protein enables specific binding to microbeads decorated with RNA-repeats of the archetypical MS2 RNA operator hairpin. Addition of the actin homologue ParM protein triggers efficient and rapid microbeads segregation via ATP-dependent ParM polymerization. Our new RNA partitioning system could be used for specific localization of mRNAs and/or the stable maintenance of self-replicating RNA vectors in various contexts such as living and artificial cells.
Predictive modeling of lower extremity injury risk in male elite youth soccer players using least absolute shrinkage and selection operator regression
(2023-01-26) Kolodziej, Mathias; Groll, Andreas; Nolte, Kevin; Willwacher, Steffen; Alt, Tobias; Schmidt, Marcus; Jaitner, Thomas
Purpose:
To (1) identify neuromuscular and biomechanical injury risk factors in elite youth soccer players and (2) assess the predictive ability of a machine learning approach.
Material and Methods:
Fifty-six elite male youth soccer players (age: 17.2 ± 1.1 years; height: 179 ± 8 cm; mass: 70.4 ± 9.2 kg) performed a 3D motion analysis, postural control testing, and strength testing. Non-contact lower extremities injuries were documented throughout 10 months. A least absolute shrinkage and selection operator (LASSO) regression model was used to identify the most important injury predictors. Predictive performance of the LASSO model was determined in a leave-one-out (LOO) prediction competition.
Results:
Twenty-three non-contact injuries were registered. The LASSO model identified concentric knee extensor peak torque, hip transversal plane moment in the single-leg drop landing task and center of pressure sway in the single-leg stance test as the three most important predictors for injury in that order. The LASSO model was able to predict injury outcomes with a likelihood of 58% and an area under the ROC curve of 0.63 (sensitivity = 35%; specificity = 79%).
Conclusion:
The three most important variables for predicting the injury outcome suggest the importance of neuromuscular and biomechanical performance measures in elite youth soccer. These preliminary results may have practical implications for future directions in injury risk screening and planning, as well as for the development of customized training programs to counteract intrinsic injury risk factors. However, the poor predictive performance of the final model confirms the challenge of predicting sports injuries, and the model must therefore be evaluated in larger samples.