Publications

Background: Short-term forecasts of infectious disease contribute to situational awareness and capacity planning. Based on best practice in other fields and recent insights in infectious disease epidemiology, one can maximise forecasts’ predictive performance by combining independent models into an ensemble. Here we report the performance of ensemble predictions of COVID-19 cases and deaths across Europe from March 2021 to March 2022. Methods: We created the European COVID-19 Forecast Hub, an online open-access platform where modellers upload weekly forecasts for 32 countries with results publicly visualised and evaluated. We created a weekly ensemble forecast from the equally-weighted average across individual models' predictive quantiles. We measured forecast accuracy using a baseline and relative Weighted Interval Score (rWIS). We retrospectively explored ensemble methods, including weighting by past performance. Results: We collected weekly forecasts from 48 models, of which we evaluated 29 models alongside the ensemble model. The ensemble had a consistently strong performance across countries over time, performing better on rWIS than 91% of forecasts for deaths (N=763 predictions from 20 models), and 83% forecasts for cases (N=886 predictions from 23 models). Performance remained stable over a 4-week horizon for death forecasts but declined with longer horizons for cases. Among ensemble methods, the most influential choice came from using a median average instead of the mean, regardless of weighting component models. Conclusions: Our results support combining independent models into an ensemble forecast to improve epidemiological predictions, and suggest that median averages yield better performance than methods based on means. We highlight that forecast consumers should place more weight on incident death forecasts than case forecasts at horizons greater than two weeks. Funding: European Commission, Ministerio de Ciencia, Innovación y Universidades, FEDER; Agència de Qualitat i Avaluació Sanitàries de Catalunya; Netzwerk Universitätsmedizin; Health Protection Research Unit; Wellcome Trust; European Centre for Disease Prevention and Control; Ministry of Science and Higher Education of Poland; Federal Ministry of Education and Research; Los Alamos National Laboratory; German Free State of Saxony; NCBiR; FISR 2020 Covid-19 I Fase; Spanish Ministry of Health / REACT-UE (FEDER); National Institutes of General Medical Sciences; Ministerio de Sanidad/ISCIII; PERISCOPE European H2020; PERISCOPE European H2021; InPresa; National Institutes of Health, NSF, US Centers for Disease Control and Prevention, Google, University of Virginia, Defense Threat Reduction Agency.

A micro segmented-flow approach was utilized for the isolation soil bacteria that can degrade synthetic polymers as polyethylene glycols (PEG) and polyacrylamide (PAM). We had been able to obtain many strains; among them, five Achromobacter spanius strains from soil samples of specific sampling sites that were connected with ancient human impacts. In addition to the characterization of community responses and isolating single strains, this microfluidic approach allowed for investigation of the susceptibility of Achromobacter spanius strains against three synthetic polymers, including PEG, PAM, and Polyvinylpyrrolidone (PVP) and two organic solvents known as 1,4-dioxane and diglyme. The small stepwise variation of effector concentrations in 500 nL droplets provides a detailed reflection of the concentration-dependent response of bacterial growth and endogenous autofluorescence activity. As a result, all five strains can use PEG600 as carbon source. Furthermore, all strains showed similar dose-response characteristics in 1,4-dioxane and diglyme. However, significantly different PAM- and PVP-tolerances were found for these strains. Samples from the surface soil of prehistorical rampart areas supplied a strain capable of degradation of PEG, PVP, and PAM. This study demonstrates on the one hand, the potential of microsegment flow for miniaturized dose-response screening studies and its ability to detect novel strains, and on the other hand, two of five isolated Achromobacter spanius strains may be useful in providing optimal growth conditions in bioremediation and biodegradation processes.

Dynamic safety-critical systems require model-based techniques and tools for their systems design. The paper presents a stochastic Petri net model of an industrial safetycritical cloud server architecture for train control. Its reliability has to be evaluated to assess tradeoffs in architecture and level of fault tolerance. Simulation methods are too slow for such rare-event problems, while numerical analysis techniques suffer from the state-space explosion problem. The paper extends a recently developed multi-trajectory simulation algorithm combining elements of simulation and numerical analysis such that it increases the accuracy of rare-event simulations within a given computation time budget. Simulation experiments have been carried out with a prototype tool.

Disease modelling has had considerable policy impact during the ongoing COVID-19 pandemic, and it is increasingly acknowledged that combining multiple models can improve the reliability of outputs. Here we report insights from ten weeks of collaborative short-term forecasting of COVID-19 in Germany and Poland (12 October-19 December 2020). The study period covers the onset of the second wave in both countries, with tightening non-pharmaceutical interventions (NPIs) and subsequently a decay (Poland) or plateau and renewed increase (Germany) in reported cases. Thirteen independent teams provided probabilistic real-time forecasts of COVID-19 cases and deaths. These were reported for lead times of one to four weeks, with evaluation focused on one- and two-week horizons, which are less affected by changing NPIs. Heterogeneity between forecasts was considerable both in terms of point predictions and forecast spread. Ensemble forecasts showed good relative performance, in particular in terms of coverage, but did not clearly dominate single-model predictions. The study was preregistered and will be followed up in future phases of the pandemic.

Similarity measurement of two probability distributions is important in many applications of statistics. Embedding such distributions into a reproducing kernel Hilbert space (RKHS) has many favorable properties. The choice of the reproducing kernel is crucial in the approach. We study this question by considering the similarity of two distributions of the same class. In particular, we investigate when the RKHS embedding is "admissible" in the sense that the distance between the embeddings should become smaller when the expectations are getting closer or when the variance is increasing to infinity. We give conditions on the widely-used translation-invariant reproducing kernels to be admissible. We also extend the study to multivariate non-symmetric Gaussian distributions.