University of Tuebingen
  • Tübingen, Germany
Recent publications
There are relatively few studies that explicitly evaluate the performance of machine learning algorithms (MLAs) such as decision trees while varying conditions like data splitting strategies and feature selection methods in digital soil mapping (DSM). Since several more powerful black-box models such as Random forest (RF) exist, regular models like the Classification and Regression Tree (CART) are least applied despite being more intelligible than the former. We demonstrate a simple yet relevant way to improve the performance of a CART model for DSM while still benefiting from its intelligibility, interpretability and intuition potential. Soil organic carbon (SOC) levels across the Czech Republic are predicted at 30 m × 30 m resolution using selected covariates coupled with respective CART models. For this work, 440 topsoils (0 – 20 cm) for the Czech Republic were retrieved from the LUCAS soil database. Regarding the distinct CART models, data splitting strategies (Random, SPlit and Conditional Latin Hypercube Sampling: cLHS) and 7 feature selection methods were varied. Meanwhile, overall model results were compared using accuracy metrics including the root mean square error (RMSE). One of the satisfactory SOC model validation results based on SPlit has a root mean square error (RMSE) of 17.30 g/kg and a coefficient of determination (R²) of 0.52. The cLHS proves robust for model data splitting. Feature selection methods including stepwise regression (SWR), recursive feature elimination (RFE) and the genetic algorithm (GA) were considered computationally efficient for identifying relevant covariates. Generally, the study demonstrates the relevance and effectiveness of varying data splitting strategies and feature selection methods for improving SOC modelling via a decision tree (CART).
In plants, the RNA-directed DNA methylation (RdDM) pathway plays a major role in establishing DNA methylation. At least some components of the RdDM machinery, including the central component AGO4, are known to concentrate in a subnuclear compartment called the Cajal body in the model plant Arabidopsis thaliana. The molecular underpinnings of Cajal body localization, however, have remained elusive so far. Here, we found that Arabidopsis AGO4 (AtAGO4) fused to GFP does not present its typical Cajal body localization, when transiently expressed in Nicotiana benthamiana. Nevertheless, the endogenous AGO4 protein from N. benthamiana shows a clear accumulation in the Cajal body. Thus, our results suggest that the Cajal body localization of AtAGO4 requires specific molecular machinery that cannot be replaced by orthologues in N. benthamiana. This study presents an experimental system that could lead to mechanistic insights into the targeting of proteins to and localization in the Cajal body in plants.
Purpose: To evaluate the long-term efficacy of combined radiotherapy (RT) and hyperthermia (HT) in a large mono-institutional cohort of breast cancer (BC) patients affected by recurrent, newly diagnosed non-resectable or high risk resected tumor. Materials and methods: Records of BC patients treated with RT + HT between 1995 and 2018 were retrospectively analyzed. RT doses of 50-70 Gy concurrent to a twice per week superficial HT were applied. For HT, a temperature between 41 and 42 °C was applied for approximately 1 h. Primary endpoint was local control (LC), secondary endpoints comprised toxicity, overall survival (OS), and progression-free survival (PFS). Results: A total of 191 patients and 196 RT + HT treatments were analyzed. In 154 cases (78.6%) RT + HT was performed for patients with recurrent BC. Among these, 93 (47.4% of the entire cohort) had received RT prior to RT + HT. Median follow up was 12.7 years. LC at 2, 5, and 10 years was 76.4, 72.8, and 69.5%, respectively. OS at 2, 5, and 10 years was 73.5, 52.3, and 35.5%, respectively. PFS at 2, 5, and 10 years was 55.6, 41, and 33.6%, respectively. Predictive factors for LC were tumor stage, distant metastases, estrogen/progesterone receptor expression, resection status and number of HT fractions. At multivariate analysis tumor stage and receptor expression were significant. No acute or late toxicities higher than grade 3 were observed. Conclusion: Combined RT + HT offers long-term high LC rates with acceptable toxicity for patients with recurrent, newly diagnosed non-resectable or resected BC at high risk of relapse.
We compared embryogenesis of five species of domesticated even-toed and one odd-toed ungulate and used a phylogenetic framework to contextualize such comparison. Organ systems that occur relatively earlier in embryogenesis generally have more time to develop and therefore are found to be more mature at birth when compared to structures that appear later in development. We hypothesized that the less mature the animals’ organs are at birth, the more they are susceptible to artificial selection. The horse had the most mature organs at birth, followed by cattle, reindeer, sheep/goat, and pig. This pattern of maturity could be observed almost during the entire development. Heterochronic shifts among species were hypothesized for when fur starts to develop. Changes in the fur coloration are one of the first observable signs of domestication and the heterochrony of this trait may be related to the effects on neural crest-derived pigment cells by artificial selection. The six ungulate species also differ in the relative duration of their weaning period and the potential extent of its artificial shortening. We put all these traits in the context of their inherited evolutionary characteristics and artificial domestication process. Related to their altriciality, carnivoran domesticates, which also belong to Scrotifera, are less mature at birth than all domesticated ungulates. Although we detected clear character correlations to life history traits, it is impossible based on the present data, to trace specific exaptations to the domestication process. We hypothesize a deep time developmental penetration of adult characters into embryogenesis.
Leaf senescence is an integral part of plant development and is driven by endogenous cues such as leaf or plant age. Developmental senescence aims to maximize the usage of carbon, nitrogen and mineral resources for growth and/or for the sake of the next generation. This requires efficient reallocation of the resources out of the senescing tissue into developing parts of the plant such as new leaves, fruits and seeds. However, premature senescence can be induced by severe and long-lasting biotic or abiotic stress conditions. It serves as an exit strategy to guarantee offspring in an unfavorable environment but is often combined with a trade-off in seed number and quality. In order to coordinate the very complex process of developmental senescence with environmental signals, highly organized networks and regulatory cues have to be in place. Reactive oxygen species, especially hydrogen peroxide (H 2 O 2 ), are involved in senescence as well as in stress signaling. Here, we want to summarize the role of H 2 O 2 as a signaling molecule in leaf senescence and shed more light on how specificity in signaling might be achieved. Altered hydrogen peroxide contents in specific compartments revealed a differential impact of H 2 O 2 produced in different compartments. Arabidopsis lines with lower H 2 O 2 levels in chloroplasts and cytoplasm point to the possibility that not the actual contents but the ratio between the two different compartments is sensed by the plant cells.
Background In Germany and the EU, most headwaters are still far from reaching a good chemical and ecological status as it is required by the European Water Framework Directive (WFD), until 2027 the latest. Particularly, in densely populated areas, impacts from municipal and industrial wastewater discharges or diffuse agricultural emissions are still a matter of concern. This also applies to the Nidda River which is considered to be in a moderate to rather poor condition. In our study, we investigated short-term and long-term consequences of anthropogenic pollution on fish health via one monitoring with caged fish (CF) and two field sampling campaigns (FF). In the CF monitoring, rainbow trout ( Oncorhynchus mykiss ) were caged for seven weeks at four selected sites along the Nidda, whereas in the FF monitoring approach, feral fish, including brown trout ( Salmo trutta f. fario ), European chub ( Leuciscus cephalus ) and stone loach ( Barbatula barbatula ) were caught in June and September 2016. Results Histopathological analyses of liver and gills were conducted, accompanied by measurements of hepatic 7-ethoxyresorufin- O -deethylase (EROD) activity to assess the cytochrome P450 (CYP1A1) function, and genotoxicity via the micronucleus assay. Caged as well as field-captured fish exhibited impaired health conditions showing lesions particularly in the liver, and a presumably overwhelmed CYP1A1 system, whereas genotoxicity was not induced. The variation between sampling sites and seasons was rather low, but two trends were recognisable: (a) liver condition was poorest around spawning season and (b) tissue integrity and EROD activity were most affected downstream of industrial dischargers. Furthermore, effects were species dependent: the generally highly sensitive S. trutta f. fario proved to be impacted most, whereas L. cephalus with its pelagic lifestyle was affected less than the benthic B. barbatula, indicating a relevant contamination of sediments. Conclusion Our results confirm the impaired ecological state of the Nidda and emphasise that a sustainable improvement of aquatic ecosystem health needs to include both water quality and sediment contamination to approach the ambitious WFD goal.
Background Clinical endpoints for upcoming therapeutic trials in frontotemporal dementia (FTD) are increasingly urgent. Cognitive composite scores are often used as endpoints but are lacking in genetic FTD. We aimed to create cognitive composite scores for genetic frontotemporal dementia (FTD) as well as recommendations for recruitment and duration in clinical trial design. Methods A standardized neuropsychological test battery covering six cognitive domains was completed by 69 C9orf72 , 41 GRN , and 28 MAPT mutation carriers with CDR® plus NACC-FTLD ≥ 0.5 and 275 controls. Logistic regression was used to identify the combination of tests that distinguished best between each mutation carrier group and controls. The composite scores were calculated from the weighted averages of test scores in the models based on the regression coefficients. Sample size estimates were calculated for individual cognitive tests and composites in a theoretical trial aimed at preventing progression from a prodromal stage (CDR® plus NACC-FTLD 0.5) to a fully symptomatic stage (CDR® plus NACC-FTLD ≥ 1). Time-to-event analysis was performed to determine how quickly mutation carriers progressed from CDR® plus NACC-FTLD = 0.5 to ≥ 1 (and therefore how long a trial would need to be). Results The results from the logistic regression analyses resulted in different composite scores for each mutation carrier group (i.e. C9orf72 , GRN , and MAPT ). The estimated sample size to detect a treatment effect was lower for composite scores than for most individual tests. A Kaplan-Meier curve showed that after 3 years, ~ 50% of individuals had converted from CDR® plus NACC-FTLD 0.5 to ≥ 1, which means that the estimated effect size needs to be halved in sample size calculations as only half of the mutation carriers would be expected to progress from CDR® plus NACC FTLD 0.5 to ≥ 1 without treatment over that time period. Discussion We created gene-specific cognitive composite scores for C9orf72 , GRN , and MAPT mutation carriers, which resulted in substantially lower estimated sample sizes to detect a treatment effect than the individual cognitive tests. The GENFI-Cog composites have potential as cognitive endpoints for upcoming clinical trials. The results from this study provide recommendations for estimating sample size and trial duration.
Background: This study aims at describing the imaging features of the metastatic presentation of clear cell renal cell carcinoma (ccRCC) in arterial (AP) and portal venous phase (PVP) of contrast-enhanced-computed-tomography (CECT) during clinical follow-up (FU) and to evaluate the necessity of a dual phase approach for metastasis detection. Methods: We identified a total of 584 patients that were diagnosed with ccRCC between January 2016 and April 2020. Inclusion criteria were histologically proven ccRCC with metastatic spread, proven by histology or interim follow-up of at least 2 years and follow-up CT examination with AP and PVP CECT including thorax/abdomen and pelvis. Exclusion criteria were defined by missing or incomplete CT-scans or lack of sufficient follow-up. CT studies of 43 patients with histologically proven ccRCCs were analyzed in retrospect. AP and PVP images were analyzed by two radiologists for metastases, two additional independent radiologists analyzed PVP images only. A 5-point Likert scale was used to evaluate the likelihood off the presence of metastasis. Imaging patterns of the metastases were analyzed visually. Results: 43 patients (16 female; mean age: 67±10 years) with recurrent ccRCC and metastatic disease were included. Three imaging patterns were observed (solid, heterogeneous or cystic metastases), which rarely exhibited calcifications (2%). All metastases showed hyperenhancement in AP and PVP. Inter-reader agreement was substantial (Fleiss' κ 0.6-0.8, p<0.001). No significant differences in sensitivity or specificity between readers (AP and PVP images vs. PVP images only) were present (79.4-85.2%, 97.1-99.6%, p ≥ 0.05). The area under the receiver-operating-characteristic (ROC) curve was between 0.901and 0.922 for all four radiologists. Conclusions: Similar rates for detection, sensitivity and specificity of metastasis and local recurrence in ccRCC were observed irrespective of using a dual-phase protocol with AP and PVP or a single PVP protocol only. Thus, a single-phase examination of PVP can be sufficient for experienced radiologists to detect metastatic disease in the follow-up of ccRCC patients.
Background Since it first started operating in 2017, the European Reference Network for Rare Neurological Diseases (ERN-RND) implemented a multi-channel communication strategy to effectively reach its target audience: healthcare professionals, patients, researchers, industry representatives and the general public. We first created a website containing useful and up to date information, followed by social media accounts. Here, the analytical data collected about the ERN-RND website and social media channels was compared (Twitter, Facebook, YouTube) during two periods: October 2018 to September 2019, and the year after the ERN-RND free educational webinars were launched: from October 2019 to September 2020. This allowed us to quantify the impact of offering a tangible product (webinars) on the communication strategy. Results The analytical data obtained from October 2018 to September 2019 and from October 2019 to September 2020 clearly shows a significant increase in traffic and followers since the launch of the ERN-RND webinars in November 2019. We also created a communication survey which was disseminated between February and June 2021. We collected responses from 61 people: 38 healthcare professionals, 11 scientists, 10 patients (advocates), 2 industry representatives, 1 patient association, 1 charity representative, 1 resident and 1 master student. Most respondents answered “webinars” as the number one reason when asked about which content they look for on the ERN-RND website. Conclusions Offering a tangible product—such as the webinars presented in this report—to a specific target group (healthcare professionals) supported our communication strategy by driving traffic to ERN-RND communication channels. It has also successfully tackled ERN-RND’s general aim: by enabling the flow of knowledge on rare neurological and movement disorders to the medical community in hospitals treating patients with these rare and complex conditions, patients ultimately benefit from improved and faster diagnosis, care, and treatment. We aim to set up similar strategies to effectively reach other or the same target groups. For healthcare professionals, organising eConsultations via the Clinical Patient Management System or disseminating standards of care such as diagnostic and therapeutic algorithms as well as clinical practice guidelines might offer potential. For the patient community, organising customised and multilingual webinars could also work.
Background Stickler syndrome (STL) is a rare, clinically and molecularly heterogeneous connective tissue disorder. Pathogenic variants occurring in a variety of genes cause STL, mainly inherited in an autosomal dominant fashion. Autosomal recessive STL is ultra-rare with only four families with biallelic COL9A3 variants reported to date. Results Here, we report three unrelated families clinically diagnosed with STL carrying different novel biallelic loss of function variants in COL9A3 . Further, we have collected COL9A3 genotype–phenotype associations from the literature. Conclusion Our report substantially expands the molecular genetics and clinical basis of autosomal recessive STL and provides an overview about allelic COL9A3 disorders.
The chemical pollution crisis severely threatens human and environmental health globally. To tackle this challenge the establishment of an overarching international science–policy body has recently been suggested. We strongly support this initiative based on the awareness that humanity has already likely left the safe operating space within planetary boundaries for novel entities including chemical pollution. Immediate action is essential and needs to be informed by sound scientific knowledge and data compiled and critically evaluated by an overarching science–policy interface body. Major challenges for such a body are (i) to foster global knowledge production on exposure, impacts and governance going beyond data-rich regions (e.g., Europe and North America), (ii) to cover the entirety of hazardous chemicals, mixtures and wastes, (iii) to follow a one-health perspective considering the risks posed by chemicals and waste on ecosystem and human health, and (iv) to strive for solution-oriented assessments based on systems thinking. Based on multiple evidence on urgent action on a global scale, we call scientists and practitioners to mobilize their scientific networks and to intensify science–policy interaction with national governments to support the negotiations on the establishment of an intergovernmental body based on scientific knowledge explaining the anticipated benefit for human and environmental health.
Leverett et al. commented on the Environmental quality standard (EQS) for diclofenac derived under the European Water Framework Directive [Leverett et al. (2021) Environ Sci Eur 33: 133 ]. They postulated that the derivation of the EQS value for diclofenac is not conducted according to the EQS Technical Guidance, but rather using data of poor reliability and relevance. Consequently, the authors suggested using their alternative derived value instead. It is to be noted that the process for the EQS derivation for diclofenac is still ongoing and not finalized, and that as a consequence, any critical analysis is very premature. In general, within the current European Commission process, EQS values proposals are derived by expert groups led by the Joint Research Centre. In the specific case for diclofenac, Leverett et al. have also been actively involved as experts. This response to Leverett et al. (2021) aims to clarify the reasoning behind the proposal from a scientific point of view and to express our concern for the lack of transparency of their position in the statement of competing interests. Indeed, the authors did not disclose their participation in the expert group for deriving the diclofenac EQS value, nor that they have direct and indirect ties to a company that markets diclofenac in Europe, Glaxo Smith & Kline plc (GSK). This amounts to a significant conflict of interest and leads to disinformation to the reader.
Purpose Molecular processes in primary osteoblasts were analyzed in response to magnetic and electric field exposure to examine its potential impact on bone healing. Methods Primary osteoblasts were exposed to a combination of a magnetic field and an additional electric field (EFMF) (20 Hz, 700 mV, 5 mT, continuous sinusoids) in vitro. mRNA- and protein-expressions were assessed during a time interval of 21 days and compared with expression data obtained from control osteoblasts. Results We observed an autonomous osteoblast differentiation process in vitro under the chosen cultivation conditions. The initial proliferative phase was characterized by a constitutively high mRNA expression of extracellular matrix proteins. Concurrent EFMF exposure resulted in significanly increased cell proliferation (fold change: 1.25) and reduced mRNA-expressions of matrix components (0.5–0.75). The following reorganization of the extracellular matrix is prerequisite for matrix mineralization and is characterised by increased Ca ²⁺ deposition (1.44). On molecular level EFMF exposure led to a significant decreased thrombospondin 1 (THBS1) mRNA- (0.81) and protein- (0.54) expression, which in turn reduced the TGFß1-dependent mRNA- (0.68) and protein- (0.5) expression of transforming growth factor beta induced (ßIG-H3) significantly, an inhibitor of endochondral ossification. Consequently, EFMF exposure stimulated the expression of genes characteristic for endochondral ossification, such as collagen type 10, A1 (1.50), osteopontin (1.50) and acellular communication network factor 3 (NOV) (1.45). Conclusions In vitro exposure of osteoblasts to EFMF supports cell differentiation and induces gene- and protein-expression patterns characteristic for endochondral ossification during bone fracture healing in vivo.
Understanding what factors predict whether an urban migrant will end up in a deprived neighbourhood or not could help prevent the exploitation of vulnerable individuals. This study leveraged pseudonymized mobile money interactions combined with cell phone data to shed light on urban migration patterns and deprivation in Tanzania. Call detail records were used to identify individuals who migrated to Dar es Salaam, Tanzania's largest city. A street survey of the city's subwards was used to determine which individuals moved to more deprived areas. t-tests showed that people who settled in poorer neighbourhoods had less money coming into their mobile money account after they moved, but not before. A machine learning approach was then utilized to predict which migrants will move to poorer areas of the city, making them arguably more vulnerable to poverty, unemployment and exploitation. Features indicating the strength and location of people's social connections in Dar es Salaam before they moved ('pull factors') were found to be most predictive, more so than traditional 'push factors' such as proxies for poverty in the migrant's source region. Supplementary information: The online version contains supplementary material available at 10.1140/epjds/s13688-022-00340-y.
Background Severe COVID-19 induced acute respiratory distress syndrome (ARDS) often requires extracorporeal membrane oxygenation (ECMO). Recent German health insurance data revealed low ICU survival rates. Patient characteristics and experience of the ECMO center may determine intensive care unit (ICU) survival. The current study aimed to identify factors affecting ICU survival of COVID-19 ECMO patients. Methods 673 COVID-19 ARDS ECMO patients treated in 26 centers between January 1st 2020 and March 22nd 2021 were included. Data on clinical characteristics, adjunct therapies, complications, and outcome were documented. Block wise logistic regression analysis was applied to identify variables associated with ICU-survival. Results Most patients were between 50 and 70 years of age. PaO 2 /FiO 2 ratio prior to ECMO was 72 mmHg (IQR: 58–99). ICU survival was 31.4%. Survival was significantly lower during the 2nd wave of the COVID-19 pandemic. A subgroup of 284 (42%) patients fulfilling modified EOLIA criteria had a higher survival (38%) ( p = 0.0014, OR 0.64 (CI 0.41–0.99)). Survival differed between low, intermediate, and high-volume centers with 20%, 30%, and 38%, respectively ( p = 0.0024). Treatment in high volume centers resulted in an odds ratio of 0.55 (CI 0.28–1.02) compared to low volume centers. Additional factors associated with survival were younger age, shorter time between intubation and ECMO initiation, BMI > 35 (compared to < 25), absence of renal replacement therapy or major bleeding/thromboembolic events. Conclusions Structural and patient-related factors, including age, comorbidities and ECMO case volume, determined the survival of COVID-19 ECMO. These factors combined with a more liberal ECMO indication during the 2nd wave may explain the reasonably overall low survival rate. Careful selection of patients and treatment in high volume ECMO centers was associated with higher odds of ICU survival. Trial registration Registered in the German Clinical Trials Register (study ID: DRKS00022964, retrospectively registered, September 7th 2020, . Graphical abstract
The Laser Interferometer Space Antenna (LISA) has the potential to reveal wonders about the fundamental theory of nature at play in the extreme gravity regime, where the gravitational interaction is both strong and dynamical. In this white paper, the Fundamental Physics Working Group of the LISA Consortium summarizes the current topics in fundamental physics where LISA observations of gravitational waves can be expected to provide key input. We provide the briefest of reviews to then delineate avenues for future research directions and to discuss connections between this working group, other working groups and the consortium work package teams. These connections must be developed for LISA to live up to its science potential in these areas.
One major challenge of longitudinal data analysis is to find an appropriate statistical model that corresponds to the theory of change and the research questions at hand. In the present article, we argue that continuous-time models are well suited to study the continuously developing constructs of primary interest in the education sciences and outline key advantages of using this type of model. Furthermore, we propose the continuous-time latent curve model with structured residuals (CT-LCM-SR) as a suitable model for many research questions in the education sciences. The CT-LCM-SR combines growth and dynamic modeling and thus provides descriptions of both trends and process dynamics. We illustrate the application of the CT-LCM-SR with data from PISA reading literacy assessment of 2000 to 2018 and provide a tutorial and annotated code for setting up the CT-LCM-SR model.
Designed to set limits to air pollutant emissions, EU Directive 2008/50/EG has pushed city administrations to strengthen their commitment to environmental protection with mixed results. However, the effectiveness of these policies remains suboptimal. Within this context, this paper provides original empirical evidence on the naturemand evolution of the debate over air pollution in the EU that helps readers understand how the interplay among discourse coalitions across levels of governance relates to the implementation of air pollution directives in Europe. To provide this evidence, we apply the discourse‑coalition framework to a multi‑level governance context by comparing debates over air pollution and examining their evolution in different contexts through the use of discourse network analysis. Our results show that the debate in the EU institutions is largely dominated by pro‑environmental coalitions, whereas discussions in Stuttgart and Berlin show different degrees of permeability to such arguments. Hence, the relative prominence of certain actors and levels of governance in the local discourse helps explain the extent to which debates at the local level align with the arguments put forward by the EU institutions. For their part, governmental organizations have a tendency to act as bridges between discourse coalitions. This novel empirical evidence provides clues that help make sense of the varying levels of implementation of EU air pollution directives at the local level. In this regard, the ability of supranational and national bodies to align their discourse with those of local organizations might facilitate a common understanding and the successful implementation of EU policies. Policymakers aiming to improve air quality at the local level may want to consider engaging actively with this debate by reinforcing the arguments of pro‑environmental coalitions. They may also want to make a special effort to ensure that the messages conveyed to supranational and national bodies are also effectively conveyed to lower levels of governance.
Institution pages aggregate content on ResearchGate related to an institution. The members listed on this page have self-identified as being affiliated with this institution. Publications listed on this page were identified by our algorithms as relating to this institution. This page was not created or approved by the institution. If you represent an institution and have questions about these pages or wish to report inaccurate content, you can contact us here.
11,529 members
Marco Skardelly
  • Department of Neurosurgery
Wolfgang Bernhard
  • Department of Neonatology
Suayib Üstün
  • Center for Plant Molecular Biology
Farid El Kasmi
  • Center for Plant Molecular Biology
Geschwister-Scholl-Platz, 72074, Tübingen, Germany