Untargeted metabolomics of COVID-19 patient serum reveals potential prognostic markers of both severity and outcome

Introduction The diagnosis of COVID-19 is normally based on the qualitative detection of viral nucleic acid sequences. Properties of the host response are not measured but are key in determining outcome. Although metabolic profiles are well suited to capture host state, most metabolomics studies are either underpowered, measure only a restricted subset of metabolites, compare infected individuals against uninfected control cohorts that are not suitably matched, or do not provide a compact predictive model. Objectives Here we provide a well-powered, untargeted metabolomics assessment of 120 COVID-19 patient samples acquired at hospital admission. The study aims to predict the patient’s infection severity (i.e., mild or severe) and potential outcome (i.e., discharged or deceased). Methods High resolution untargeted UHPLC-MS/MS analysis was performed on patient serum using both positive and negative ionization modes. A subset of 20 intermediary metabolites predictive of severity or outcome were selected based on univariate statistical significance and a multiple predictor Bayesian logistic regression model was created. Results The predictors were selected for their relevant biological function and include deoxycytidine and ureidopropionate (indirectly reflecting viral load), kynurenine (reflecting host inflammatory response), and multiple short chain acylcarnitines (energy metabolism) among others. Currently, this approach predicts outcome and severity with a Monte Carlo cross validated area under the ROC curve of 0.792 (SD 0.09) and 0.793 (SD 0.08), respectively. A blind validation study on an additional 90 patients predicted outcome and severity at ROC AUC of 0.83 (CI 0.74–0.91) and 0.76 (CI 0.67–0.86). Conclusion Prognostic tests based on the markers discussed in this paper could allow improvement in the planning of COVID-19 patient treatment.

Frugivore gut passage increases seed germination: an updated meta-analysis

Many plants rely on animal mutualists for reproduction. Quantifying how animal mutualists impact plant performance provides a foundation for modelling how change in animal communities affects the composition and functioning of plant communities. We performed a meta-analysis of 2539 experiments, 6 times more than the last comprehensive meta-analysis, examining how gut passage by frugivores influences seed germination. We simultaneously analyzed multiple predictor variables related to study methodology, location, and frugivore identity to disentangle methodological from ecological impacts on effect sizes. We found that gut passage by birds, fish, reptiles, bats, primates, and other mammals on average increased seed germination, but that the magnitude varied across vertebrate groups. The positive effects of gut passage were largely explained by the de-inhibitory effects of pulp removal rather than by the scarification of seed tissues. Some previous studies and meta-analyses that found no effect of gut passage only tested scarification or did not distinguish between these tests of scarification and pulp removal. We found that, for a typical fleshy-fruited plant species, the lack of gut passage reduces germination by 60%. From an evolutionary perspective, this indicates a large risk associated with reliance on animal mutualists that is balanced against the benefits of animal-mediated seed dispersal. From a conservation perspective, this highlights the potential for large demographic consequences of frugivore declines on plant populations. Our database and findings advance quantitative predictions for the role of fruit-frugivore interactions in shaping plant communities in the Anthropocene.

Non-spatial and Spatial Linear Regression

This chapter gives an introduction to the workhorse of quantitative statistical analysis, linear regression analysis, assuming minimal background knowledge of the reader. We give a broad overview of linear regression analysis using one predictor variable and then turn to regression with multiple predictor variables and key assumptions, which segues into regression analysis of areal units that include spatial dependence. Throughout we use the statistical programming environment R, and we try to summarize the most important challenges that an applied researcher will face. As this is just an introduction to the topic, we provide references to sources that are highly recommended for any researcher who aims to understand or apply (spatial) linear regression analysis.

Untargeted metabolomics of COVID-19 patient serum reveals potential prognostic markers of both severity and outcome

The diagnosis of COVID-19 is normally based on the qualitative detection of viral nucleic acid sequences. Properties of the host response are not measured but are key in determining outcome. Although metabolic profiles are well suited to capture host state, existing metabolomics studies are either underpowered, measure only a restricted subset of metabolites (‘targeted metabolomics’), compare infected individuals against uninfected control cohorts that are not suitably matched, or do not provide a compact predictive model.We here provide a well-powered, untargeted metabolomics assessment of 120 COVID-19 patient samples acquired at hospital admission. The study aims to predict patient’s infection severity (i.e. mild or severe) and potential outcome (i.e. discharged or deceased).High resolution untargeted LC-MS/MS analysis was performed on patient serum using both positive and negative ionization. A subset of 20 intermediary metabolites predictive of severity or outcome were selected based on univariate statistical significance and a multiple predictor Bayesian logistic regression model. The predictors were selected for their relevant biological function and include cytosine (reflecting viral load), kynurenine (reflecting host inflammatory response), nicotinuric acid, and multiple short chain acylcarnitines (energy metabolism) among others.Currently, this approach predicts outcome and severity with a Monte Carlo cross validated area under the ROC curve of 0.792 (SD 0.09) and 0.793 (SD 0.08), respectively. Prognostic tests based on the markers discussed in this paper could allow improvement in the planning of COVID-19 patient treatment.

Community structure and functional genes in radionuclide contaminated soils in Chernobyl and Fukushima

ABSTRACT Chernobyl and Fukushima were subjected to radionuclide (RN) contamination that has led to environmental problems. In order to explore the ability of microorganisms to survive in these environments, we used a combined 16S rRNA and metagenomic approach to describe the prokaryotic community structure and metabolic potential over a gradient of RN concentrations (137Cs 1680–0.4 and 90Sr 209.1–1.9 kBq kg−1) in soil samples. The taxonomic results showed that samples with low 137Cs content (37.8–0.4 kBq kg−1) from Fukushima and Chernobyl clustered together. In order to determine the effect of soil chemical parameters such as organic carbon (OC), Cesium-137 (137Cs) and Strontium-90 (90Sr) on the functional potential of microbial communities, multiple predictor model analysis using piecewiseSEM was carried out on Chernobyl soil metagenomes. The model identified 46 genes that were correlated to these parameters of which most have previously been described as mechanisms used by microorganisms under stress conditions. This study provides a baseline taxonomic and metagenomic dataset for Fukushima and Chernobyl, respectively, including physical and chemical characteristics. Our results pave the way for evaluating the possible RN selective pressure that might contribute to shaping microbial community structure and their functions in contaminated soils.

A machine learning method for subgroup analysis of randomized controlled trials

We developed a machine learning method for subgroup analyses of randomized controlled trials (RCT), and applied it to the results of the SPRINT RCT for treatment of hypertension. To date, the subgroup analyses mostly focused on detecting associations between certain factors and outcome, in the hope that the results will point out biologically (for example, carriers of a certain mutation) or clinically (for example, smokers) distinct subgroups with different outcomes. This seldom worked in the sense of re-launching the intervention for the detected subgroup only and successfully treating it. In contrast, we propose an empirical and general method to develop a predictive multivariate classifier using the RCT outcomes and baseline data. The classifier identifies patients likely to benefit from the intervention, is not limited to a single factor of interest, and is ready for validation in a subsequent pivotal trial. We believe this approach has a better chance of succeeding in identifying the relevant subgroups because of increased accuracy made possible by the use of multiple predictor variables, and opportunity to use advanced machine learning. The method effectiveness is demonstrated by the analysis of the SPRINT trial.

ANALISIS FINANCIAL DISTRESS DAN BEBERAPA VARIABEL PREDIKTORPADA SEKTOR PERBANKAN DI INDONESIA

The Purpose of this study is (1) to analyze financial distress as predictor variables of financial difficulty levels banking which is listing on the Stock Exchange (BEI), and (2) determine the financial ratios, including CAR, ROE, NPL, NIM, LDR, ROA, BOPO, and the primary reserve as predictor variables of financial distress in banking sector which is listing on the BEI. The population of this study were 38 banks which is listed on the BEI until December 31th, 2014. Number of samples were 30 banks by purposive sampling technique. The analysis tool of study used Logit regression because of dependent variable was dummy variable and the independent variables were a combination of metric and non-metric. The results showed (1) banks were listed on the BEI has good performance (financial difficulties at low levels), the analysis result of financial distress by using multiple predictor variables, such as CAR, ROE, NPL, NIM, LDR, BOPO and reserves' primary GWM has good average value where is level of accuracy (97.2) in classifying the bank's financial difficulties; (2) The hypothesis testing showed a number of variables such as CAR, ROE, NIM, LDR, and BOPO were not significant as a predictor of financial distress in banking where listed on the BEI. However, NPL and reserves' primary GWM were significant variable as predictor of financial distress in banking where listed on the BEI. Keywords: Reserves' primary GWM, Non-Performing Loan, Financial Distress