Ludwig-Maximilians-Universität München
Open Access LMU ( Ludwig-Maximilians-Univ. München)Not a member yet
40914 research outputs found
Sort by
Flow-type controls on tributary alluvial fan formation along the Andes (18-34°S)
Get PDFDebris flows, locally known as ‘aluviones’ or ‘huaicos’, in Andean tributary alluvial fans form distinctive facies associations that can be classified into High-density Flows (HdF) and Low-Density Flows (LdF) based on sediment-to-water ratios and transitions from highly dense, viscous flows to more diluted ones. This distinction, based on ground evidence and remotely sensed debris flow identification, establishes the first equivalence between field-based and optical satellite imagery observations using Google Earth. By analysing the activity of tributary alluvial fans over the past 20 years with open-access optical satellite imagery, we provide new insights into fan evolution and the extent to which lateral sediment inputs impact sediment transfer along axial river valleys of the Andes.
Our observations reveal that large-volume debris flows, typically associated with HdF, contribute to fan expansion and aggradation. In contrast, more diluted flows -ranging from hyperconcentrated to fluvial flows-promote fan destruction through incision of the feeder channel. Fans dominated by LdF facies associations often exhibit incision and progradation, forming new lobes at the fan toe. These lobes are frequently reworked by the main channel due to the limited sediment supply. Whether tributary fan sedimentation and progradation influence the main channel or if sediments are buffered on the fans largely depends on the characteristics of the flows and on the original topography of the fan. Accordingly, classifying debris flow surges into two distinct facies associations (AF1 and AF2, corresponding to HdF and LdF, respectively) enhances our understanding of fan dynamics and their influence on axial valleys at regional scale. This study underscores the importance of sedimentological flows characteristics in governing fan evolution, influencing both fan development across a broad latitudinal range (18-34°S) and sedimentary signal propagation along the Andes Cordillera. Moreover, the findings have significant implications for national debris flow hazard mitigation efforts and aligns with global strategies outlined in the Sendai Framework for Disaster Risk Reduction, promoting resilience and improved risk management in developing regions
Probabilistic scoring lists for interpretable machine learning
Get PDFA scoring system is a simple decision model that checks a set of features, adds a certain number of points to a total score for each feature that is satisfied, and finally makes a decision by comparing the total score to a threshold. Scoring systems have a long history of active use in safety-critical domains such as healthcare and justice, where they provide guidance for making objective and accurate decisions. Given their genuine interpretability, the idea of learning scoring systems from data is obviously appealing from the perspective of explainable AI. In this paper, we propose a practically motivated extension of scoring systems called probabilistic scoring lists (PSL), as well as a method for learning PSLs from data. Instead of making a deterministic decision, a PSL represents uncertainty in the form of probability distributions, or, more generally, probability intervals. Moreover, in the spirit of decision lists, a PSL evaluates features one by one and stops as soon as a decision can be made with enough confidence. To evaluate our approach, we conduct case studies in the medical domain and on standard benchmark data
Expanding the Allelic and Clinical Heterogeneity of Movement Disorders Linked to Defects of Mitochondrial Adenosine Triphosphate Synthase
Get PDFBackground: Defects of mitochondrial ATP synthase (ATPase) represent an emerging, yet incompletely understood group of neurodevelopmental diseases with abnormal movements.
Objective: The aim of this study was to redefine the phenotypic and mutational spectrum of movement disorders linked to the ATPase subunit-encoding genes ATP5F1A and ATP5F1B.
Methods: We recruited regionally distant patients who had been genome or exome sequenced. Fibroblast cultures from two patients were established to perform RNA sequencing, immunoblotting, mass spectrometry–based high-throughput quantitative proteomics, and ATPase activity assays. In silico three-dimensional missense variant modeling was performed.
Results: We identified a patient with developmental delay, myoclonic dystonia, and spasticity who carried a heterozygous frameshift c.1404del (p.Glu469Serfs*3) variant in ATP5F1A. The patient's cells exhibited significant reductions in ATP5F1A mRNA, underexpression of the α-subunit of ATPase in association with other aberrantly expressed ATPase components, and compromised ATPase activity. In addition, a novel deleterious heterozygous ATP5F1A missense c.1252G>A (p.Gly418Arg) variant was discovered, shared by three patients from two families with hereditary spastic paraplegia (HSP). This variant mapped to a functionally important intersubunit communication site. A third heterozygous variant, c.1074+1G>T, affected a canonical donor splice site of ATP5F1B and resulted in exon skipping with significantly diminished ATP5F1B mRNA levels, as well as impaired ATPase activity. The associated phenotype consisted of cerebral palsy (CP) with prominent generalized dystonia.
Conclusions: Our data confirm and expand the role of dominant ATP5F1A and ATP5F1B variants in neurodevelopmental movement disorders. ATP5F1A/ATP5F1B-related ATPase diseases should be considered as a cause of dystonia, HSP, and CP. © 2025 The Author(s). Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society
Probabilistic neural operators for functional uncertainty quantification
Get PDFNeural operators aim to approximate the solution operator of a system of differential equations purely from data. They have shown immense success in modeling complex dynamical systems across various domains. However, the occurrence of uncertainties inherent in both model and data has so far rarely been taken into account\textemdash{}a critical limitation in complex, chaotic systems such as weather forecasting. In this paper, we introduce the probabilistic neural operator (PNO), a framework for learning probability distributions over the output function space of neural operators. PNO extends neural operators with generative modeling based on strictly proper scoring rules, integrating uncertainty information directly into the training process. We provide a theoretical justification for the approach and demonstrate improved performance in quantifying uncertainty across different domains and with respect to different baselines. Furthermore, PNO requires minimal adjustment to existing architectures, shows improved performance for most probabilistic prediction tasks, and leads to well-calibrated predictive distributions and adequate uncertainty representations even for long dynamical trajectories. Implementing our approach into large-scale models for physical applications can lead to improvements in corresponding uncertainty quantification and extreme event identification, ultimately leading to a deeper understanding of the prediction of such surrogate models
Human Footprint and Forest Disturbance Reduce Space Use of Brown Bears (Ursus arctos) Across Europe
Get PDFThree-quarters of the planet's land surface has been altered by humans, with consequences for animal ecology, movements and related ecosystem functioning. Species often occupy wide geographical ranges with contrasting human disturbance and environmental conditions, yet, limited data availability across species' ranges has constrained our understanding of how human pressure and resource availability jointly shape intraspecific variation of animal space use. Leveraging a unique dataset of 758 annual GPS movement trajectories from 375 brown bears (Ursus arctos) across the species' range in Europe, we investigated the effects of human pressure (i.e., human footprint index), resource availability and predictability, forest cover and disturbance, and area-based conservation measures on brown bear space use. We quantified space use at different spatiotemporal scales during the growing season (May–September): home range size; representing general space requirements, 10-day long-distance displacement distances, and routine 1-day displacement distances. We found large intraspecific variation in brown bear space use across all scales, which was profoundly affected by human footprint index, vegetation productivity, and recent forest disturbances creating opportunity for resource pulses. Bears occupied smaller home ranges and moved less in more anthropized landscapes and in areas with higher resource availability and predictability. Forest disturbances reduced space use while contiguous forest cover promoted longer daily movements. The amount of strictly protected and roadless areas within bear home ranges was too small to affect space use. Anthropized landscapes may hinder the expansion of small and isolated populations, such as the Apennine and Pyrenean, and obstruct population connectivity, for example between the Dinaric Pindos population and the Alpine or Carpathian population. Our findings call for actions to maintain bear movements across landscapes with high human footprint, for example by maintaining forest integrity, to support viable bear populations and their ecosystem functions
Printed CsMg–ZnO ETLs achieve over 9 % efficiency in PbS quantum dot solar cells
Get PDFZinc oxide (ZnO) is a key electron transport layer (ETL) material in next-generation lead sulfide (PbS) colloidal quantum dot solar cells (CQDSCs) due to its high transparency, strong exciton binding energy, and good electron mobility. Here, we demonstrate a scalable doctor-blading printing protocol for ZnO ETLs that integrates dual defect passivation with magnesium (Mg2⁺) and caesium (Cs⁺) and employs solvent engineering to achieve uniform, defect-minimized films. Using a ternary solvent blend (methanol, chloroform, and 2-methoxyethanol) optimizes the ink's viscosity and boiling point, preventing particle migration and ensuring full substrate coverage. Our modified ZnO ink leads to improved crystallinity, smoother surfaces, and reduced trap states, boosting the fill factor (FF) and short-circuit current (Jsc). Consequently, we achieve a power conversion efficiency increase from 5.98 % to 9.53 % using a printed CsMg-ZnO film. Notably, 80 % of dual-doped devices exceeded 7.5 % efficiency, demonstrating high reproducibility and reliability. This performance enhancement underscores the effectiveness of dual metal ion treatment and solvent engineering strategies in overcoming printability challenges. Moreover, the compatibility of our approach with low-temperature processing and established coating techniques paves the way for seamless integration into large-scale manufacturing, bringing PbS CQDSCs closer to commercial viability
FOLFIRI with cetuximab or bevacizumab in RAS wild-type metastatic colorectal cancer: Refining first-line treatment selection by combining clinical parameters
No full textBackground
Primary tumor sidedness (PTS) with discrimination of left-sided (LC) and right-sided tumors (RC) guides patient selection for targeted first-line therapy in RAS wild-type (RAS-WT) metastatic colorectal cancer (mCRC). This study assessed the hypothesis whether considering PTS with additional clinical parameters better predicts the treatment benefit of targeted first-line treatment.
Methods
In FIRE-3, first-line treatment with folinic acid, fluorouracil and irinotecan (FOLFIRI) plus cetuximab (FOLFIRI/Cet) was compared to FOLFIRI plus bevacizumab (FOLFIRI/Bev) in patients with RAS-WT mCRC and unresectable metastasis. We evaluated whether combining PTS with number of metastatic sites (NOM), liver-limited disease status (LLD), age, sex, or carcinoembryonic antigen level (CEA) better predicts treatment benefit regarding overall survival (OS). Here, Cox regression models with second-order interactions were applied. Further, the results were validated by policy learning and Lasso regression analysis.
Findings
Among 400 RAS-WT mCRC patients, combining PTS with LLD status in a Cox regression model outperformed PTS alone for predicted treatment benefit (P = 0·005; c‑index=0·603). Significant OS benefit from FOLFIRI/Cet over FOLFIRI/Bev was observed in LC/non-LLD patients (HR=0·62; 95 %-confidence interval [CI]=0·46–0·82; P = 0·002), but mitigated in LC/LLD patients (HR=0·83; 95 %-CI=0·53–1·29; P = 0·400). In RC/non-LLD patients, FOLFIRI/Bev demonstrated a significant OS advantage over FOLFIRI/Cet (HR=2·09; 95 %‑CI=1·20–3·63; P = 0·010). However, RC/LLD patients showed potential benefit from FOLFIRI/Cet, though not statistically significant (HR=0·59; 95 %-CI=0·25–1·39; P = 0·218).
Interpretation
Incorporating PTS and LLD status might improve selection of targeted first-line treatment in RAS-WT mCRC patients. FOLFIRI/Cet appears to be particularly beneficial for LC/non-LLD patients with mitigated benefit in patients with LC/LLD. In contrast, FOLFIRI/Bev is significantly favoured over FOLFIRI/Cet in patients with RC/non-LLD. Notably, RC/LLD patients may still benefit from anti-EGFR therapy despite right-sided primary tumor. These results are hypothesis-generating and warrant further validation
