Quantifiable identification of flow-limited ventilator dyssynchrony with the deformed lung ventilator model

BACKGROUND

Ventilator dyssynchrony (VD) can worsen lung injury and is challenging to detect and quantify due to the complex variability in the dyssynchronous breaths. While machine learning (ML) approaches are useful for automating VD detection from the ventilator waveform data, scalable severity quantification and its association with pathogenesis and ventilator mechanics remain challenging.

OBJECTIVE

We develop a systematic framework to quantify pathophysiological features observed in ventilator waveform signals such that they can be used to create feature-based severity stratification of VD breaths.

METHODS

A mathematical model was developed to represent the pressure and volume waveforms of individual breaths in a feature-based parametric form. Model estimates of respiratory effort strength were used to assess the severity of flow-limited (FL)-VD breaths compared to normal breaths. A total of 93,007 breath waveforms from 13 patients were analyzed.

RESULTS

A novel model-defined continuous severity marker was developed and used to estimate breath phenotypes of FL-VD breaths. The phenotypes had a predictive accuracy of over 97% with respect to the previously developed ML-VD identification algorithm. To understand the incidence of FL-VD breaths and their association with the patient state, these phenotypes were further successfully correlated with ventilator-measured parameters and electronic health records.

CONCLUSION

This work provides a computational pipeline to identify and quantify the severity of FL-VD breaths and paves the way for a large-scale study of VD causes and effects. This approach has direct application to clinical practice and in meaningful knowledge extraction from the ventilator waveform data.

Statistical approach for radioactivity detection: A brief review

Radioactivity detection is a major research and development priority for many practical applications. Amongst the various technical challenges in this field is the need to carry out accurate low-level radioactivity measurements in the presence of a large fluctuations in the natural radiation background, while reducing the false alarm rates. The task becomes even more harder with high detection limits under low signal-to-background ratios. A detection method based on the statistical inference, following either a frequentist or a Bayesian paradigm, adopted to overcome these challenges as well as to ensure a reliable and accurate diagnosis with a competitive tradeoff between sensitivity, specificity and response time. With this respect, several research studies, addressing a range of applications from decommissioning and dismantling to homeland security, have been proposed. Our main goal in this paper is to present a succinct survey of these studies based on a frequentist and Bayesian approaches used to decision-making, uncertainty and risk evaluation, in the context of radioactive detection. In this prospect, a theoretical background of statistical frequentist and Bayesian inferences was presented. Then, a comparative study of both approaches was performed to determine the optimal approach in regards to accuracy and pros/cons. A case of study for low-level radioactivity detection in nuclear decommissioning operations was provided to validate the optimal approach. Results proved the efficiency and usefulness of Bayesian approach against frequentist one with respect to the most challenging scenarios in radiation detection applications.

Three-phase hierarchical model-based and hybrid inference

Global commitments to mitigating climate change and halting biodiversity loss require reliable information about Earth's ecosystems. Increasingly, such information is obtained from multiple sources of remotely sensed data combined with data acquired in the field. This new wealth of data poses challenges regarding the combination of different data sources to derive the required information and assess uncertainties. In this article, we show how predictors and their variances can be derived when hierarchically nested models are applied. Previous studies have developed methods for cases involving two modeling steps, such as biomass prediction relying on tree-level allometric models and models linking plot-level field data with remotely sensed data. This study extends the analysis to cases involving three modeling steps to cover new important applications. The additional step might involve an intermediate model, linking field and remotely sensed data available from a small sample, for making predictions that are subsequently used for training a final prediction model based on remotely sensed data:•In cases where the data in the final step are available wall-to-wall, we denote the approach three-phase hierarchical model-based inference (3pHMB),•In cases where the data in the final step are available as a probability sample, we denote the approach three-phase hierarchical hybrid inference (3pHHY).

Neural correlates of valence-dependent belief and value updating during uncertainty reduction: An fNIRS study

Selective use of new information is crucial for adaptive decision-making. Combining a gamble bidding task with assessing cortical responses using functional near-infrared spectroscopy (fNIRS), we investigated potential effects of information valence on behavioral and neural processes of belief and value updating during uncertainty reduction in young adults. By modeling changes in the participants' expressed subjective values using a Bayesian model, we dissociated processes of (i) updating beliefs about statistical properties of the gamble, (ii) updating values of a gamble based on new information about its winning probabilities, as well as (iii) expectancy violation. The results showed that participants used new information to update their beliefs and values about the gambles in a quasi-optimal manner, as reflected in the selective updating only in situations with reducible uncertainty. Furthermore, their updating was valence-dependent: information indicating an increase in winning probability was underweighted, whereas information about a decrease in winning probability was updated in good agreement with predictions of the Bayesian decision theory. Results of model-based and moderation analyses showed that this valence-dependent asymmetry was associated with a distinct contribution of expectancy violation, besides belief updating, to value updating after experiencing new positive information regarding winning probabilities. In line with the behavioral results, we replicated previous findings showing involvements of frontoparietal brain regions in the different components of updating. Furthermore, this study provided novel results suggesting a valence-dependent recruitment of brain regions. Individuals with stronger oxyhemoglobin responses during value updating was more in line with predictions of the Bayesian model while integrating new information that indicates an increase in winning probability. Taken together, this study provides first results showing expectancy violation as a contributing factor to sub-optimal valence-dependent updating during uncertainty reduction and suggests limitations of normative Bayesian decision theory.

Statistical modeling of gut microbiota for personalized health status monitoring

BACKGROUND

The gut microbiome is closely associated with health status, and any microbiota dysbiosis could considerably impact the host's health. In addition, many active consortium projects have generated many reference datasets available for large-scale retrospective research. However, a comprehensive monitoring framework that analyzes health status and quantitatively present bacteria-to-health contribution has not been thoroughly investigated.

METHODS

We systematically developed a statistical monitoring diagram for personalized health status prediction and analysis. Our framework comprises three elements: (1) a statistical monitoring model was established, the health index was constructed, and the health boundary was defined; (2) healthy patterns were identified among healthy people and analyzed using contrast learning; (3) the contribution of each bacterium to the health index of the diseased population was analyzed. Furthermore, we investigated disease proximity using the contribution spectrum and discovered multiple multi-disease-related targets.

RESULTS

We demonstrated and evaluated the effectiveness of the proposed monitoring framework for tracking personalized health status through comprehensive real-data analysis using the multi-study cohort and another validation cohort. A statistical monitoring model was developed based on 92 microbial taxa. In both the discovery and validation sets, our approach achieved balanced accuracies of 0.7132 and 0.7026, and AUC of 0.80 and 0.76, respectively. Four health patterns were identified in healthy populations, highlighting variations in species composition and metabolic function across these patterns. Furthermore, a reasonable correlation was found between the proposed health index and host physiological indicators, diversity, and functional redundancy. The health index significantly correlated with Shannon diversity ([Formula: see text]) and species richness ([Formula: see text]) in the healthy samples. However, in samples from individuals with diseases, the health index significantly correlated with age ([Formula: see text]), species richness ([Formula: see text]), and functional redundancy ([Formula: see text]). Personalized diagnosis is achieved by analyzing the contribution of each bacterium to the health index. We identified high-contribution species shared across multiple diseases by analyzing the contribution spectrum of these diseases.

CONCLUSIONS

Our research revealed that the proposed monitoring framework could promote a deep understanding of healthy microbiomes and unhealthy variations and served as a bridge toward individualized therapy target discovery and precise modulation. Video Abstract.

MSstats Version 4.0: Statistical Analyses of Quantitative Mass Spectrometry-Based Proteomic Experiments with Chromatography-Based Quantification at Scale

The MSstats R-Bioconductor family of packages is widely used for statistical analyses of quantitative bottom-up mass spectrometry-based proteomic experiments to detect differentially abundant proteins. It is applicable to a variety of experimental designs and data acquisition strategies and is compatible with many data processing tools used to identify and quantify spectral features. In the face of ever-increasing complexities of experiments and data processing strategies, the core package of the family, with the same name MSstats, has undergone a series of substantial updates. Its new version MSstats v4.0 improves the usability, versatility, and accuracy of statistical methodology, and the usage of computational resources. New converters integrate the output of upstream processing tools directly with MSstats, requiring less manual work by the user. The package's statistical models have been updated to a more robust workflow. Finally, MSstats' code has been substantially refactored to improve memory use and computation speed. Here we detail these updates, highlighting methodological differences between the new and old versions. An empirical comparison of MSstats v4.0 to its previous implementations, as well as to the packages MSqRob and DEqMS, on controlled mixtures and biological experiments demonstrated a stronger performance and better usability of MSstats v4.0 as compared to existing methods.

Reducing the complexity of high-dimensional environmental data: An analytical framework using LASSO with considerations of confounding for statistical inference

PURPOSE

Frameworks for selecting exposures in high-dimensional environmental datasets, while considering confounding, are lacking. We present a two-step approach for exposure selection with subsequent confounder adjustment for statistical inference.

METHODS

We measured cognitive ability in 338 children using the Woodcock-Muñoz General Intellectual Ability (GIA) score, and potential associated features across several environmental domains. Initially, 111 variables theoretically associated with GIA score were introduced into a Least Absolute Shrinkage and Selection Operator (LASSO) in a 50% feature selection subsample. Effect estimates for selected features were subsequently modeled in linear regressions in a 50% inference (hold out) subsample, first adjusting for sex and age and later for covariates selected via directed acyclic graphs (DAGs). All models were adjusted for clustering by school.

RESULTS

Of the 15 LASSO selected variables, eleven were not associated with GIA score following our inference modeling approach. Four variables were associated with GIA scores, including: serum ferritin adjusted for inflammation (inversely), mother's IQ (positively), father's education (positively), and hours per day the child works on homework (positively). Serum ferritin was not in the expected direction.

CONCLUSIONS

Our two-step approach moves high-dimensional feature selection a step further by incorporating DAG-based confounder adjustment for statistical inference.

Visual search and childhood vision impairment: A GAMLSS-oriented multiverse analysis approach

The aim of this report was to analyze reaction times and accuracy in children with a vision impairment performing a feature-based visual search task using a multiverse statistical approach. The search task consisted of set sizes 4, 16, and 24, consisting of distractors (circle) and a target (ellipse) that were presented randomly to school-aged individuals with or without a vision impairment. Interactions and main effects of key variables relating to reaction times and accuracy were analyzed via a novel statistical method blending GAMLSS (generalized additive models for location, scale, and shape) and distributional regression trees. Reaction times for the target-present and target-absent conditions were significantly slower in the vision impairment group with increasing set sizes (p < .001). Female participants were significantly slower than were males for set sizes 16 and 24 in the target-absent condition (p < .001), with male participants being significantly slower than females in the target-present condition (p < .001). Accuracy was only significantly worse (p = .03) for participants less than 14 years of age for the target-absent condition with set sizes 16 and 24. There was a positive association between binocular visual acuity and search time (p < .001). The application of GAMLSS with distributional regression trees to the analysis of visual search data may provide further insights into underlying factors affecting search performance in case-control studies where psychological or physical differences may influence visual search outcomes.

An evaluation of homeostatic plasticity for ecosystems using an analytical data science approach

The natural world is constantly changing, and planetary boundaries are issuing severe warnings about biodiversity and cycles of carbon, nitrogen, and phosphorus. In other views, social problems such as global warming and food shortages are spreading to various fields. These seemingly unrelated issues are closely related, but it can be said that understanding them in an integrated manner is still a step away. However, progress in analytical technologies has been recognized in various fields and, from a microscopic perspective, with the development of instruments including next-generation sequencers (NGS), nuclear magnetic resonance (NMR), gas chromatography-mass spectrometry (GC/MS), and liquid chromatography-mass spectrometry (LC/MS), various forms of molecular information such as genome data, microflora structure, metabolome, proteome, and lipidome can be obtained. The development of new technology has made it possible to obtain molecular information in a variety of forms. From a macroscopic perspective, the development of environmental analytical instruments and environmental measurement facilities such as satellites, drones, observation ships, and semiconductor censors has increased the data availability for various environmental factors. Based on these background, the role of computational science is to provide a mechanism for integrating and understanding these seemingly disparate data sets. This review describes machine learning and the need for structural equations and statistical causal inference of these data to solve these problems. In addition to introducing actual examples of how these technologies can be utilized, we will discuss how to use these technologies to implement environmentally friendly technologies in society.

The multilevel hierarchical data EM-algorithm. Applications to discrete-time Markov chain epidemic models

The theory of multilevel hierarchical data Expectation Maximization (EM)-algorithm is introduced via discrete time Markov chain (DTMC) epidemic models. A general model for a multilevel hierarchical discrete data is derived. The observed sample Y in the system is a stochastic incomplete data, and the missing data Z exhibits a multilevel hierarchical data structure. The EM-algorithm to find ML-estimates for parameters in the stochastic system is derived. Applications of the EM-algorithm are exhibited in the two DTMC models, to find ML-estimates of the system parameters. Numerical results are given for influenza epidemics in the state of Georgia (GA), USA.

To curb research misreporting, replace significance and confidence by compatibility: A Preventive Medicine Golden Jubilee article

It is well known that the statistical analyses in health-science and medical journals are frequently misleading or even wrong. Despite many decades of reform efforts by hundreds of scientists and statisticians, attempts to fix the problem by avoiding obvious error and encouraging good practice have not altered this basic situation. Statistical teaching and reporting remain mired in damaging yet editorially enforced jargon of "significance", "confidence", and imbalanced focus on null (no-effect or "nil") hypotheses, leading to flawed attempts to simplify descriptions of results in ordinary terms. A positive development amidst all this has been the introduction of interval estimates alongside or in place of significance tests and P-values, but intervals have been beset by similar misinterpretations. Attempts to remedy this situation by calling for replacement of traditional statistics with competitors (such as pure-likelihood or Bayesian methods) have had little impact. Thus, rather than ban or replace P-values or confidence intervals, we propose to replace traditional jargon with more accurate and modest ordinary-language labels that describe these statistics as measures of compatibility between data and hypotheses or models, which have long been in use in the statistical modeling literature. Such descriptions emphasize the full range of possibilities compatible with observations. Additionally, a simple transform of the P-value called the surprisal or S-value provides a sense of how much or how little information the data supply against those possibilities. We illustrate these reforms using some examples from a highly charged topic: trials of ivermectin treatment for Covid-19.

Choosing multiple linear regressions for weather-based crop yield prediction with ABSOLUT v1.2 applied to the districts of Germany

ABSOLUT v1.2 is an adaptive algorithm that uses correlations between time-aggregated weather variables and crop yields for yield prediction. In contrast to conventional regression-based yield prediction methods, a very broad range of possible input features and their combinations are exhaustively tested for maximum explanatory power. Weather variables such as temperature, precipitation, and sunshine duration are aggregated over different seasonal time periods preceding the harvest to 45 potential input features per original variable. In a first step, this large set of features is reduced to those aggregates very probably holding explanatory power for observed yields. The second, computationally demanding step evaluates predictions for all districts with all of their possible combinations. Step three selects those combinations of weather features that showed the highest predictive power across districts. Finally, the district-specific best performing regressions among these are used for actual prediction, and the results are spatially aggregated. To evaluate the new approach, ABSOLUT v1.2 is applied to predict the yields of silage maize, winter wheat, and other major crops in Germany based on two decades of data from about 300 districts. It turned out to be absolutely crucial to not only make out-of-sample predictions (solely based on data excluding the target year to predict) but to also consequently separate training and testing years in the process of feature selection. Otherwise, the prediction accuracy would be over-estimated by far. The question arises whether performances claimed for other statistical modelling examples are often upward-biased through input variable selection disregarding the out-of-sample principle.

A logical analysis of null hypothesis significance testing using popular terminology

Background

Null Hypothesis Significance Testing (NHST) has been well criticised over the years yet remains a pillar of statistical inference. Although NHST is well described in terms of statistical models, most textbooks for non-statisticians present the null and alternative hypotheses (H₀ and H_A, respectively) in terms of differences between groups such as (μ₁ = μ₂) and (μ₁ ≠ μ₂) and H_A is often stated to be the research hypothesis. Here we use propositional calculus to analyse the internal logic of NHST when couched in this popular terminology. The testable H₀ is determined by analysing the scope and limits of the P-value and the test statistic’s probability distribution curve.

Results

We propose a minimum axiom set NHST in which it is taken as axiomatic that H₀ is rejected if P-value< α. Using the common scenario of the comparison of the means of two sample groups as an example, the testable H₀ is {(μ₁ = μ₂) and [(\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\overline{x}$$\end{document}x¯₁ ≠ \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\overline{x}$$\end{document}x¯₂) due to chance alone]}. The H₀ and H_A pair should be exhaustive to avoid false dichotomies. This entails that H_A is ¬{(μ₁ = μ₂) and [(\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\overline{x}$$\end{document}x¯₁ ≠ \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\overline{x}$$\end{document}x¯₂) due to chance alone]}, rather than the research hypothesis (H_T). To see the relationship between H_A and H_T, H_A can be rewritten as the disjunction H_A: ({(μ₁ = μ₂) ∧ [(\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\overline{x}$$\end{document}x¯₁ ≠ \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\overline{x}$$\end{document}x¯₂) not due to chance alone]} ∨ {(μ₁ ≠ μ₂) ∧ [\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(\overline{x}$$\end{document}(x¯₁ ≠ \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\overline{x}$$\end{document}x¯₂) not due to (μ₁ ≠ μ₂) alone]} ∨ {(μ₁ ≠ μ₂) ∧ [(\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\overline{\boldsymbol{x}}$$\end{document}x¯₁≠\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\overline{\boldsymbol{x}}$$\end{document}x¯₂) due to (μ₁ ≠ μ₂) alone]}). This reveals that H_T (the last disjunct in bold) is just one possibility within H_A. It is only by adding premises to NHST that H_T or other conclusions can be reached.

Conclusions

Using this popular terminology for NHST, analysis shows that the definitions of H₀ and H_A differ from those found in textbooks. In this framework, achieving a statistically significant result only justifies the broad conclusion that the results are not due to chance alone, not that the research hypothesis is true. More transparency is needed concerning the premises added to NHST to rig particular conclusions such as H_T. There are also ramifications for the interpretation of Type I and II errors, as well as power, which do not specifically refer to H_T as claimed by texts.

Bayesian causal inference: A unifying neuroscience theory

Understanding of the brain and the principles governing neural processing requires theories that are parsimonious, can account for a diverse set of phenomena, and can make testable predictions. Here, we review the theory of Bayesian causal inference, which has been tested, refined, and extended in a variety of tasks in humans and other primates by several research groups. Bayesian causal inference is normative and has explained human behavior in a vast number of tasks including unisensory and multisensory perceptual tasks, sensorimotor, and motor tasks, and has accounted for counter-intuitive findings. The theory has made novel predictions that have been tested and confirmed empirically, and recent studies have started to map its algorithms and neural implementation in the human brain. The parsimony, the diversity of the phenomena that the theory has explained, and its illuminating brain function at all three of Marr's levels of analysis make Bayesian causal inference a strong neuroscience theory. This also highlights the importance of collaborative and multi-disciplinary research for the development of new theories in neuroscience.

Markov genealogy processes

We construct a family of genealogy-valued Markov processes that are induced by a continuous-time Markov population process. We derive exact expressions for the likelihood of a given genealogy conditional on the history of the underlying population process. These lead to a nonlinear filtering equation which can be used to design efficient Monte Carlo inference algorithms. We demonstrate these calculations with several examples. Existing full-information approaches for phylodynamic inference are special cases of the theory.

Misinterpretations of P-values and statistical tests persists among researchers and professionals working with statistics and epidemiology

BACKGROUND

The aim was to investigate inferences of statistically significant test results among persons with more or less statistical education and research experience.

METHODS

A total of 75 doctoral students and 64 statisticians/epidemiologist responded to a web questionnaire about inferences of statistically significant findings. Participants were asked about their education and research experience, and also whether a 'statistically significant' test result (P = 0.024, α-level 0.05) could be inferred as proof or probability statements about the truth or falsehood of the null hypothesis (H₀) and the alternative hypothesis (H₁).

RESULTS

Almost all participants reported having a university degree, and among statisticians/epidemiologist, most reported having a university degree in statistics and were working professionally with statistics. Overall, 9.4% of statisticians/epidemiologist and 24.0% of doctoral students responded that the statistically significant finding proved that H₀ is not true, and 73.4% of statisticians/epidemiologists and 53.3% of doctoral students responded that the statistically significant finding indicated that H₀ is improbable. Corresponding numbers about inferences about the alternative hypothesis (H₁) were 12.0% and 6.2% about proving H₁ being true and 62.7 and 62.5% for the conclusion that H₁ is probable. Correct inferences to both questions, which is that a statistically significant finding cannot be inferred as either proof or a measure of a hypothesis' probability, were given by 10.7% of doctoral students and 12.5% of statisticians/epidemiologists.

CONCLUSIONS

Misinterpretation of P-values and statistically significant test results persists also among persons who have substantial statistical education and who work professionally with statistics.

Identifying Developmental Patterns in Structured Plant Phenotyping Data

Technological breakthroughs concerning both sensors and robotized plant phenotyping platforms have totally renewed the plant phenotyping paradigm in the last two decades. This has impacted both the nature and the throughput of data with the availability of data at high-throughput from the tissular to the whole plant scale. Sensor outputs often take the form of 2D or 3D images or time series of such images from which traits are extracted while organ shapes, shoot or root system architectures can be deduced. Despite this change of paradigm, many phenotyping studies often ignore the structure of the plant and therefore loose the information conveyed by the temporal and spatial patterns emerging from this structure. The developmental patterns of plants often take the form of succession of well-differentiated phases, stages or zones depending on the temporal, spatial or topological indexing of data. This entails the use of hierarchical statistical models for their identification.The objective here is to show potential approaches for analyzing structured plant phenotyping data using state-of-the-art methods combining probabilistic modeling, statistical inference and pattern recognition. This approach is illustrated using five different examples at various scales that combine temporal and topological index parameters, and development and growth variables obtained using prospective or retrospective measurements.

adabmDCA: adaptive Boltzmann machine learning for biological sequences

Background

Boltzmann machines are energy-based models that have been shown to provide an accurate statistical description of domains of evolutionary-related protein and RNA families. They are parametrized in terms of local biases accounting for residue conservation, and pairwise terms to model epistatic coevolution between residues. From the model parameters, it is possible to extract an accurate prediction of the three-dimensional contact map of the target domain. More recently, the accuracy of these models has been also assessed in terms of their ability in predicting mutational effects and generating in silico functional sequences.

Results

Our adaptive implementation of Boltzmann machine learning, adabmDCA, can be generally applied to both protein and RNA families and accomplishes several learning set-ups, depending on the complexity of the input data and on the user requirements. The code is fully available at https://github.com/anna-pa-m/adabmDCA. As an example, we have performed the learning of three Boltzmann machines modeling the Kunitz and Beta-lactamase2 protein domains and TPP-riboswitch RNA domain.

Conclusions

The models learned by adabmDCA are comparable to those obtained by state-of-the-art techniques for this task, in terms of the quality of the inferred contact map as well as of the synthetically generated sequences. In addition, the code implements both equilibrium and out-of-equilibrium learning, which allows for an accurate and lossless training when the equilibrium one is prohibitive in terms of computational time, and allows for pruning irrelevant parameters using an information-based criterion.

Towards a pragmatic use of statistics in ecology

Although null hypothesis testing (NHT) is the primary method for analyzing data in many natural sciences, it has been increasingly criticized. Recently, approaches based on information theory (IT) have become popular and were held by many to be superior because it enables researchers to properly assess the strength of the evidence that data provide for competing hypotheses. Many studies have compared IT and NHT in the context of model selection and stepwise regression, but a systematic comparison of the most basic uses of statistics by ecologists is still lacking. We used computer simulations to compare how both approaches perform in four basic test designs (t-test, ANOVA, correlation tests, and multiple linear regression). Performance was measured by the proportion of simulated samples for which each method provided the correct conclusion (power), the proportion of detected effects with a wrong sign (S-error), and the mean ratio of the estimated effect to the true effect (M-error). We also checked if the p-value from significance tests correlated to a measure of strength of evidence, the Akaike weight. In general both methods performed equally well. The concordance is explained by the monotonic relationship between p-values and evidence weights in simple designs, which agree with analytic results. Our results show that researchers can agree on the conclusions drawn from a data set even when they are using different statistical approaches. By focusing on the practical consequences of inferences, such a pragmatic view of statistics can promote insightful dialogue among researchers on how to find a common ground from different pieces of evidence. A less dogmatic view of statistical inference can also help to broaden the debate about the role of statistics in science to the entire path that leads from a research hypothesis to a statistical hypothesis.

Biasing the input: A yoked-scientist demonstration of the distorting effects of optional stopping on Bayesian inference

Prior work by Michael R. Dougherty and colleagues (Yu et al., 2014) shows that when a scientist monitors the p value during data collection and uses a critical p as the signal to stop collecting data, the resulting p is distorted due to Type I error-rate inflation. They argued similarly that the use of a critical Bayes factor (BF_(crit)) for stopping distorts the obtained Bayes factor (BF), a position that has met with controversy. The present paper clarified that when BF_(crit) is used as a stopping criterion, the sample becomes biased in that data consistent with large effects have a greater chance to be included than do other data, thus biasing the input to Bayesian inference. We report simulations of yoked pairs of scientists in which Scientist A uses BF_(crit) to optionally stop, while Scientist B, sampling from the same population, stops when A stops. Thus, optional stopping is compared not to a hypothetical in which no stopping occurs, but to a situation in which B stops for reasons unrelated to the characteristics of B's sample. The results indicated that optional stopping biased the input for Bayesian inference. We also simulated the use of effect-size stabilization as a stopping criterion and found no bias in that case.

Estimating the generation interval and inferring the latent period of COVID-19 from the contact tracing data

The coronavirus disease 2019 (COVID-19) emerged by end of 2019, and became a serious public health threat globally in less than half a year. The generation interval and latent period, though both are of importance in understanding the features of COVID-19 transmission, are difficult to observe, and thus they can rarely be learnt from surveillance data empirically. In this study, we develop a likelihood framework to estimate the generation interval and incubation period simultaneously by using the contact tracing data of COVID-19 cases, and infer the pre-symptomatic transmission proportion and latent period thereafter. We estimate the mean of incubation period at 6.8 days (95 %CI: 6.2, 7.5) and SD at 4.1 days (95 %CI: 3.7, 4.8), and the mean of generation interval at 6.7 days (95 %CI: 5.4, 7.6) and SD at 1.8 days (95 %CI: 0.3, 3.8). The basic reproduction number is estimated ranging from 1.9 to 3.6, and there are 49.8 % (95 %CI: 33.3, 71.5) of the secondary COVID-19 infections likely due to pre-symptomatic transmission. Using the best estimates of model parameters, we further infer the mean latent period at 3.3 days (95 %CI: 0.2, 7.9). Our findings highlight the importance of both isolation for symptomatic cases, and for the pre-symptomatic and asymptomatic cases.

Why optional stopping can be a problem for Bayesians

Recently, optional stopping has been a subject of debate in the Bayesian psychology community. Rouder (Psychonomic Bulletin & Review 21(2), 301-308, 2014) argues that optional stopping is no problem for Bayesians, and even recommends the use of optional stopping in practice, as do (Wagenmakers, Wetzels, Borsboom, van der Maas & Kievit, Perspectives on Psychological Science 7, 627-633, 2012). This article addresses the question of whether optional stopping is problematic for Bayesian methods, and specifies under which circumstances and in which sense it is and is not. By slightly varying and extending Rouder's (Psychonomic Bulletin & Review 21(2), 301-308, 2014) experiments, we illustrate that, as soon as the parameters of interest are equipped with default or pragmatic priors-which means, in most practical applications of Bayes factor hypothesis testing-resilience to optional stopping can break down. We distinguish between three types of default priors, each having their own specific issues with optional stopping, ranging from no-problem-at-all (type 0 priors) to quite severe (type II priors).

Searching for universal model of amyloid signaling motifs using probabilistic context-free grammars

Background

Amyloid signaling motifs are a class of protein motifs which share basic structural and functional features despite the lack of clear sequence homology. They are hard to detect in large sequence databases either with the alignment-based profile methods (due to short length and diversity) or with generic amyloid- and prion-finding tools (due to insufficient discriminative power). We propose to address the challenge with a machine learning grammatical model capable of generalizing over diverse collections of unaligned yet related motifs.

Results

First, we introduce and test improvements to our probabilistic context-free grammar framework for protein sequences that allow for inferring more sophisticated models achieving high sensitivity at low false positive rates. Then, we infer universal grammars for a collection of recently identified bacterial amyloid signaling motifs and demonstrate that the method is capable of generalizing by successfully searching for related motifs in fungi. The results are compared to available alternative methods. Finally, we conduct spectroscopy and staining analyses of selected peptides to verify their structural and functional relationship.

Conclusions

While the profile HMMs remain the method of choice for modeling homologous sets of sequences, PCFGs seem more suitable for building meta-family descriptors and extrapolating beyond the seed sample.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-021-04139-y.

How to use likelihood ratios to interpret evidence from randomized trials

OBJECTIVE

The likelihood ratio is a method for assessing evidence regarding two simple statistical hypotheses. Its interpretation is simple - for example, a value of 10 means that the first hypothesis is 10 times as strongly supported by the data as the second. A method is shown for deriving likelihood ratios from published trial reports.

STUDY DESIGN

The likelihood ratio compares two hypotheses in light of data: that a new treatment is effective, at a specified level (alternate hypothesis: for instance, the hazard ratio equals 0.7), and that it is not (null hypothesis: the hazard ratio equals 1). The result of the trial is summarised by the test statistic z (ie, the estimated treatment effect divided by its standard error). The expected value of z is 0 under the null hypothesis, and A under the alternate hypothesis. The logarithm of the likelihood ratio is given by z·A - A²/2. The values of A and z can be derived from the alternate hypothesis used for sample size computation, and from the observed treatment effect and its standard error or confidence interval.

RESULTS

Examples are given of trials that yielded strong or moderate evidence in favor of the alternate hypothesis, and of a trial that favored the null hypothesis. The resulting likelihood ratios are applied to initial beliefs about the hypotheses to obtain posterior beliefs.

CONCLUSIONS

The likelihood ratio is a simple and easily understandable method for assessing evidence in data about two competing a priori hypotheses.

Inflexible Updating of the Self-Other Divide During a Social Context in Autism: Psychophysical, Electrophysiological, and Neural Network Modeling Evidence

BACKGROUND

Autism spectrum disorder (ASD) affects many aspects of life, from social interactions to (multi)sensory processing. Similarly, the condition expresses at a variety of levels of description, from genetics to neural circuits and interpersonal behavior. We attempt to bridge between domains and levels of description by detailing the behavioral, electrophysiological, and putative neural network basis of peripersonal space (PPS) updating in ASD during a social context, given that the encoding of this space relies on appropriate multisensory integration, is malleable by social context, and is thought to delineate the boundary between the self and others.

METHODS

Fifty (20 male/30 female) young adults, either diagnosed with ASD or age- and sex-matched individuals, took part in a visuotactile reaction time task indexing PPS, while high-density electroencephalography was continuously recorded. Neural network modeling was performed in silico.

RESULTS

Multisensory psychophysics demonstrates that while PPS in neurotypical individuals shrinks in the presence of others-as to "give space"-this does not occur in ASD. Likewise, electroencephalography recordings suggest that multisensory integration is altered by social context in neurotypical individuals but not in individuals with ASD. Finally, a biologically plausible neural network model shows, as a proof of principle, that PPS updating may be inflexible in ASD owing to the altered excitatory/inhibitory balance that characterizes neural circuits in animal models of ASD.

CONCLUSIONS

Findings are conceptually in line with recent statistical inference accounts, suggesting diminished flexibility in ASD, and further these observations by suggesting within an example relevant for social cognition that such inflexibility may be due to excitatory/inhibitory imbalances.

Computational framework for investigating predictive processing in auditory perception

BACKGROUND

The brain tracks sound sources as they evolve in time, collecting contextual information to predict future sensory inputs. Previous work in predictive coding typically focuses on the perception of predictable stimuli, leaving the implementation of these same neural processes in more complex, real-world environments containing randomness and uncertainty up for debate.

NEW METHOD

To facilitate investigation into the perception of less tightly-controlled listening scenarios, we present a computational model as a tool to ask targeted questions about the underlying predictive processes that connect complex sensory inputs to listener behavior and neural responses. In the modeling framework, observed sound features (e.g. pitch) are tracked sequentially using Bayesian inference. Sufficient statistics are inferred from past observations at multiple time scales and used to make predictions about future observation while tracking the statistical structure of the sensory input.

RESULTS

Facets of the model are discussed in terms of their application to perceptual research, and examples taken from real-world audio demonstrate the model's flexibility to capture a variety of statistical structures along various perceptual dimensions.

COMPARISON WITH EXISTING METHODS

Previous models are often targeted toward interpreting a particular experimental paradigm (e.g., oddball paradigm), perceptual dimension (e.g., pitch processing), or task (e.g., speech segregation), thus limiting their ability to generalize to other domains. The presented model is designed as a flexible and practical tool for broad application.

CONCLUSION

The model is presented as a general framework for generating new hypotheses and guiding investigation into the neural processes underlying predictive coding of complex scenes.

Identifying clinical and biochemical phenotypes in acute respiratory distress syndrome secondary to coronavirus disease-2019

BACKGROUND

Acute respiratory distress syndrome (ARDS) secondary to coronavirus disease-2019 (COVID-19) is characterized by substantial heterogeneity in clinical, biochemical, and physiological characteristics. However, the pathophysiology of severe COVID-19 infection is poorly understood. Previous studies established clinical and biological phenotypes among classical ARDS cohorts, with important therapeutic implications. The phenotypic profile of COVID-19 associated ARDS remains unknown.

METHODS

We used latent class modeling via a multivariate mixture model to identify phenotypes from clinical and biochemical data collected from 263 patients admitted to Massachusetts General Hospital intensive care unit with COVID-19-associated ARDS between March 13 and August 2, 2020.

FINDINGS

We identified two distinct phenotypes of COVID-19-associated ARDS, with substantial differences in biochemical profiles despite minimal differences in respiratory dynamics. The minority phenotype (class 2, n = 70, 26·6%) demonstrated increased markers of coagulopathy, with mild relative hyper-inflammation and dramatically increased markers of end-organ dysfunction (e.g., creatinine, troponin). The odds of 28-day mortality among the class 2 phenotype was more than double that of the class 1 phenotype (40·0% vs.· 23·3%, OR = 2·2, 95% CI [1·2, 3·9]).

INTERPRETATION

We identified distinct phenotypic profiles in COVID-19 associated ARDS, with little variation according to respiratory physiology but with important variation according to systemic and extra-pulmonary markers. Phenotypic identity was highly associated with short-term mortality. The class 2 phenotype exhibited prominent signatures of coagulopathy, suggesting that vascular dysfunction may play an important role in the clinical progression of severe COVID-19-related disease.

Estimates of the mean difference in orthopaedic randomized trials: obligatory yet obscure

Background

Randomized controlled trials in orthopaedics are powered to mainly find large effect sizes. A possible discrepancy between the estimated and the real mean difference is a challenge for statistical inference based on p-values. We explored the justifications of the mean difference estimates used in power calculations. The assessment of distribution of observations in the primary outcome and the possibility of ceiling effects were also assessed.

Methods

Systematic review of the randomized controlled trials with power calculations in eight clinical orthopaedic journals published between 2016 and 2019. Trials with one continuous primary outcome and 1:1 allocation were eligible. Rationales and references for the mean difference estimate were recorded from the Methods sections. The possibility of ceiling effect was addressed by the assessment of the weighted mean and standard deviation of the primary outcome and its elaboration in the Discussion section of each RCT where available.

Results

264 trials were included in this study. Of these, 108 (41 %) trials provided some rationale or reference for the mean difference estimate. The most common rationales or references for the estimate of mean difference were minimal clinical important difference (16 %), observational studies on the same subject (8 %) and the ‘clinical relevance’ of the authors (6 %). In a third of the trials, the weighted mean plus 1 standard deviation of the primary outcome reached over the best value in the patient-reported outcome measure scale, indicating the possibility of ceiling effect in the outcome.

Conclusions

The chosen mean difference estimates in power calculations are rarely properly justified in orthopaedic trials. In general, trials with a patient-reported outcome measure as the primary outcome do not assess or report the possibility of the ceiling effect in the primary outcome or elaborate further in the Discussion section.

Informal versus formal judgment of statistical models: The case of normality assumptions

Researchers sometimes use informal judgment for statistical model diagnostics and assumption checking. Informal judgment might seem more desirable than formal judgment because of a paradox: Formal hypothesis tests of assumptions appear to become less useful as sample size increases. We suggest that this paradox can be resolved by evaluating both formal and informal statistical judgment via a simplified signal detection framework. In 4 studies, we used this approach to compare informal judgments of normality diagnostic graphs (histograms, Q-Q plots, and P-P plots) to the performance of several formal tests (Shapiro-Wilk test, Kolmogorov-Smirnov test, etc.). Participants judged whether or not graphs of sample data came from a normal population (Experiments 1-2) or whether or not from a population close enough to normal for a parametric test to be more powerful than a nonparametric one (Experiments 3-4). Across all experiments, participants' informal judgments showed lower discriminability than did formal hypothesis tests. This pattern occurred even after participants were given 400 training trials with feedback, a financial incentive, and ecologically valid distribution shapes. The discriminability advantage of formal normality tests led to slightly more powerful follow-up tests (parametric vs. nonparametric). Overall, the framework used here suggests that formal model diagnostics may be more desirable than informal ones.

Robust Likelihood-Based Approach for Automated Optimization and Uncertainty Analysis of Toxicokinetic-Toxicodynamic Models

Toxicokinetic-toxicodynamic (TKTD) models offer a mechanistic understanding of individual-level toxicity over time and allow for meaningful extrapolations from laboratory tests to exposure conditions in the field. Thereby, they hold great potential for ecotoxicological studies, both in a regulatory context as well as for basic research. In contrast to mechanistic effect models at higher levels of biological organization, TKTD models can be, and generally are, parameterized by fitting them to data (results from toxicity tests). Fitting models comes with a range of statistical and numerical challenges, which may hamper the application of TKTD models in a practical setting. Especially in the context of environmental risk assessment, there is a need for robust and user-friendly software tools to automatically extract the best-fitting model parameters and quantify their uncertainty from any data set. The study presents a general outline for TKTD model analysis, rooted in likelihood-based ("frequentist") inference. The general outline is followed by a presentation of the specific algorithm that has been implemented into software for the robust and automated analysis of toxicity data for survival. However, the presented approach is more broadly applicable to low-dimensional problems. Integr Environ Assess Manag 2021;17:388-397. © 2020 SETAC.

Inferencing superspreading potential using zero-truncated negative binomial model: exemplification with COVID-19

Background

In infectious disease transmission dynamics, the high heterogeneity in individual infectiousness indicates that few index cases generate large numbers of secondary cases, which is commonly known as superspreading events. The heterogeneity in transmission can be measured by describing the distribution of the number of secondary cases as a negative binomial (NB) distribution with dispersion parameter, k. However, such inference framework usually neglects the under-ascertainment of sporadic cases, which are those without known epidemiological link and considered as independent clusters of size one, and this may potentially bias the estimates.

Methods

In this study, we adopt a zero-truncated likelihood-based framework to estimate k. We evaluate the estimation performance by using stochastic simulations, and compare it with the baseline non-truncated version. We exemplify the analytical framework with three contact tracing datasets of COVID-19.

Results

We demonstrate that the estimation bias exists when the under-ascertainment of index cases with 0 secondary case occurs, and the zero-truncated inference overcomes this problem and yields a less biased estimator of k. We find that the k of COVID-19 is inferred at 0.32 (95%CI: 0.15, 0.64), which appears slightly smaller than many previous estimates. We provide the simulation codes applying the inference framework in this study.

Conclusions

The zero-truncated framework is recommended for less biased transmission heterogeneity estimates. These findings highlight the importance of individual-specific case management strategies to mitigate COVID-19 pandemic by lowering the transmission risks of potential super-spreaders with priority.

Recent Developments in Species Sensitivity Distribution Modeling

The species sensitivity distribution (SSD) is a statistical approach that is used to estimate either the concentration of a chemical that is hazardous to no more than x% of all species (the HCx) or the proportion of species potentially affected by a given concentration of a chemical. Despite a significant body of published research and critical reviews over the past 20 yr aimed at improving the methodology, the fundamentals remain unchanged. Although there have been some recent suggestions for improvements to SSD methods in the literature, in general, few of these suggestions have been formally adopted. Furthermore, critics of the approach can rightly point to the fact that differences in technical implementation can lead to marked differences in results, thereby undermining confidence in SSD approaches. Despite the limitations, SSDs remain a practical tool and, until a demonstrably better inferential framework is available, developments and enhancements to conventional SSD practice will and should continue. We therefore believe the time has come for the scientific community to decide how it wants SSD methods to evolve. The present study summarizes the current status of, and elaborates on several recent developments for, SSD methods, specifically, model averaging, multimodality, and software development. We also consider future directions with respect to the use of SSDs, with the ultimate aim of helping to facilitate greater international collaboration and, potentially, greater harmonization of SSD methods. Environ Toxicol Chem 2021;40:293-308. © 2020 SETAC.

A novel, unbiased approach to evaluating subsequent search misses in dual target visual search

Research in radiology and visual cognition suggest that finding one target during visual search may result in increased misses for a second target, an effect known as subsequent search misses (SSM). Here, we demonstrate that the common method of calculating second-target detection performance is biased and could produce spurious SSM effects. We describe the source of that bias and document factors that influence its magnitude. We use a modification of signal-detection theory to develop a novel, unbiased method of calculating the expected value for dual-target performance under the null hypothesis. We then apply our novel method to two of our data sets that showed modest SSM effects when calculated in the traditional manner. Our correction reduced the effect size to the point that there was no longer a significant SSM effect. We then applied our method to a published data set that had a larger effect size when calculated using the traditional calculation as well as when using an alternative calculation that was recently proposed to account for biases in the traditional method. We find that both the traditional method and the recently proposed alternative substantially overestimate the magnitude of the SSM effect in these data, but a significant SSM effect persisted even with our calculation. We recommend that future SSM studies use our method to ensure accurate effect-size estimates, and suggest that the method be applied to reanalyze published results, particularly those with small effect sizes, to rule out the possibility that they were spurious.

Learning Lessons on Reproducibility and Replicability in Large Scale Genome-Wide Association Studies

Reproducibility and replicability play a pivotal role in science. The article reflects on reproducibility and replicability as they figure in large scale genome-wide association studies. Overall, we emphasize the importance of enhancing data reproducibility, analysis reproducibility, and result replicability. We make recommendations pertaining to the development of study designs that address 1) batch effects and selection bias, 2) the incorporation of discrete discovery and replication phases, and 3) the procurement of a large sample size. We emphasize the importance of systematic and transparent data generation, processing, and quality control pipelines, as well as a rigorous field-specific standardized analysis protocol, We offer guidance with respect to collaborative frameworks, open access analysis tools, and software, and the use of supporting mandates, infrastructure, and repositories for data and resource sharing. Finally, we identify the role of incentives and culture in fueling the production of reproducible and replicable research through partnerships of researchers, funding agencies, and journals.

To avoid the noncausal association between environmental factor and COVID-19 when using aggregated data: Simulation-based counterexamples for demonstration

In the infectious disease epidemiology, the association between an independent factor and disease incidence (or death) counts may fail to infer the association with disease transmission (or mortality risk). To explore the underlying role of environmental factors in the course of COVID-19 epidemic, the importance of following the epidemiological metric's definition and systematic analytical procedures are highlighted. Cautiousness needs to be taken when understanding the outcome association based on the aggregated data, and overinterpretation should be avoided. The existing analytical approaches to address the inferential failure mentioned in this study are also discussed.

Estimating and testing an index of bias attributable to composite outcomes in comparative studies

OBJECTIVES

This study aimed to develop an index to evaluate the bias attributable to composite outcomes (BACOs) in comparative clinical studies.

STUDY DESIGN AND SETTING

The author defined the BACO index as the ratio of the logarithm of the association measure (e.g., relative risk) of the composite outcome to that of its most relevant component endpoint (e.g., mortality). Methods to calculate the confidence intervals and test the null hypotheses (BACO index = 1) were described and applied in systematically selected clinical trials. Two other preselected trials were included as "positive controls" for being examples of primary composite outcomes disregarded because of inconsistency with the treatment effect on mortality.

RESULTS

The BACO index values different from 1 were classified according to whether the use of composite outcomes overestimated (BACO index >1), underestimated (BACO index between 0 and <1), or inverted (BACO index <0) the association between exposure and prognosis. In 3 of 23 systematically selected trials and the two positive controls, the BACO indices were significantly lower than 1 (P < 0.005).

CONCLUSION

BACO index can warn that the composite outcome association is stronger, weaker, or even opposite than that of its most critical component.

Many meta-analyses of rare events in the Cochrane Database of Systematic Reviews were underpowered

BACKGROUND AND OBJECTIVE

Meta-analysis is a statistical method with the ability to increase the power for statistical inference, while it may still face the problem of being underpowered. In this study, we investigated the power to detect certain true effects for published meta-analyses of rare events.

METHODS

We extracted data from the Cochrane Database of Systematic Reviews for meta-analyses of rare events from January 2003 to May 2018. We retrospectively estimated the power to detect a 10-50% relative risk reduction (RRR) of eligible meta-analyses. The proportion of meta-analyses achieved a sufficient power (≥0.8) were estimated.

RESULTS

We identified 4,177 meta-analyses. The median power to detect 10%, 30%, and 50% RRR were 0.06 (interquartile range [IQR]: 0.05 to 0.06), 0.08 (IQR: 0.06 to 0.15), and 0.17 (IQR: 0.10 to 0.42), respectively); the corresponding proportion of meta-analyses that reached sufficient power were 0.32%, 3.68%, and 11.81%. Meta-analyses incorporating data from more studies had higher probability to achieve a sufficient power (rate ratio = 2.49, 95% CI: 1.76, 3.52, P < 0.001).

CONCLUSION

Most of the meta-analyses of rare events in Cochrane systematic reviews were underpowered. Future meta-analysis of rare events should report the power of the results to support informative conclusions.

Principles and Reporting of Bayesian Trials

Bayesian clinical trials are becoming popular owing to their adaptive, flexible, and versatile nature. Such trials typically require specification of the prior distribution and construction of the likelihood function; subsequently, inference is made on the basis of the posterior distribution. In comparison with frequentist trial designs, there are less established guidelines on how to report Bayesian trials. We provide a general overview on key components of the design, conduct, and analysis of Bayesian trials and elaborate on the reporting guidelines dos and don'ts.

Unsupervised inference approach to facial attractiveness

The perception of facial attractiveness is a complex phenomenon which depends on how the observer perceives not only individual facial features, but also their mutual influence and interplay. In the machine learning community, this problem is typically tackled as a problem of regression of the subject-averaged rating assigned to natural faces. However, it has been conjectured that this approach does not capture the complexity of the phenomenon. It has recently been shown that different human subjects can navigate the face-space and "sculpt" their preferred modification of a reference facial portrait. Here we present an unsupervised inference study of the set of sculpted facial vectors in such experiments. We first infer minimal, interpretable and accurate probabilistic models (through Maximum Entropy and artificial neural networks) of the preferred facial variations, that encode the inter-subject variance. The application of such generative models to the supervised classification of the gender of the subject that sculpted the face reveals that it may be predicted with astonishingly high accuracy. We observe that the classification accuracy improves by increasing the order of the non-linear effective interaction. This suggests that the cognitive mechanisms related to facial discrimination in the brain do not involve the positions of single facial landmarks only, but mainly the mutual influence of couples, and even triplets and quadruplets of landmarks. Furthermore, the high prediction accuracy of the subjects' gender suggests that much relevant information regarding the subjects may influence (and be elicited from) their facial preference criteria, in agreement with the multiple motive theory of attractiveness proposed in previous works.

Trialling Meta-Research in Comparative Cognition: Claims and Statistical Inference in Animal Physical Cognition

Scientific disciplines face concerns about replicability and statistical inference, and these concerns are also relevant in animal cognition research. This paper presents a first attempt to assess how researchers make and publish claims about animal physical cognition, and the statistical inferences they use to support them. We surveyed 116 published experiments from 63 papers on physical cognition, covering 43 different species. The most common tasks in our sample were trap-tube tasks (14 papers), other tool use tasks (13 papers), means-end understanding and string-pulling tasks (11 papers), object choice and object permanence tasks (9 papers) and access tasks (5 papers). This sample is not representative of the full scope of physical cognition research; however, it does provide data on the types of statistical design and publication decisions researchers have adopted. Across the 116 experiments, the median sample size was 7. Depending on the definitions we used, we estimated that between 44% and 59% of our sample of papers made positive claims about animals' physical cognitive abilities, between 24% and 46% made inconclusive claims, and between 10% and 17% made negative claims. Several failures of animals to pass physical cognition tasks were reported. Although our measures had low inter-observer reliability, these findings show that negative results can and have been published in the field. However, publication bias is still present, and consistent with this, we observed a drop in the frequency of p-values above .05. This suggests that some non-significant results have not been published. More promisingly, we found that researchers are likely making many correct statistical inferences at the individual-level. The strength of evidence of statistical effects at the group-level was weaker, and its p-value distribution was consistent with some effect sizes being overestimated. Studies such as ours can form part of a wider investigation into statistical reliability in comparative cognition. However, future work should focus on developing the validity and reliability of the measurements they use, and we offer some starting points.

How feasible is it to abandon statistical significance? A reflection based on a short survey

Background

There is a growing trend in using the “statistically significant” term in the scientific literature. However, harsh criticism of this concept motivated the recommendation to withdraw its use of scientific publications. We aimed to validate the support and the feasibility of adherence to this recommendation, among researchers having declared in favor of removing the statistical significance.

Methods

We surveyed signatories of an article published that defended this recommendation, to validate their opinion and ask them about how likely they will retire the concept of statistical significance.

Results

We obtained 151 responses which confirmed the support for the mentioned publication in aspects such as the adequate interpretation of the p-value, the degree of agreement, and the motivations to sign it. However, there was a wide distribution of answers about how likely are they to use the concept of “statistical significance” in future publications. About 42% declared being neutral, or that would likely use it again. We described arguments referred by several signatories and discussed aspects to be considered in the interpretation of research results.

Conclusions

The responses obtained from a proportion of signatories validated their declared position against the use of statistical significance. However, even in this group, the full application of this recommendation does not seem feasible. The arguments related to the inappropriate use of statistical tests should promote more education among researchers and users of scientific evidence.

Resampling-based methods for biologists

Ecological data often violate common assumptions of traditional parametric statistics (e.g., that residuals are Normally distributed, have constant variance, and cases are independent). Modern statistical methods are well equipped to handle these complications, but they can be challenging for non-statisticians to understand and implement. Rather than default to increasingly complex statistical methods, resampling-based methods can sometimes provide an alternative method for performing statistical inference, while also facilitating a deeper understanding of foundational concepts in frequentist statistics (e.g., sampling distributions, confidence intervals, p-values). Using simple examples and case studies, we demonstrate how resampling-based methods can help elucidate core statistical concepts and provide alternative methods for tackling challenging problems across a broad range of ecological applications.

Exclusion of studies with no events in both arms in meta-analysis impacted the conclusions

OBJECTIVES

Classical meta-analyses routinely treated studies with no events in both arms noninformative and excluded them from analyses. This study assessed whether such studies contain information and have an influence on the conclusions of meta-analyses.

STUDY DESIGN AND SETTING

We collected meta-analyses of binary outcomes with at least one study having no events in both arms from Cochrane systematic reviews (2003-2018). We used the generalized linear mixed model to reanalyze these meta-analyses by two approaches: one including studies with no events in both arms and one excluding such studies. The magnitude and direction of odds ratio (OR), P value, and width of 95% confidence interval (CI) were compared. A simulation study was conducted to examine the robustness of results.

RESULTS

We identified 442 meta-analyses. In comparing paired meta-analyses that included studies with no events in both arms vs. those not, 8 (1.80%) resulted in different directions on OR; 41 (9.28%) altered conclusions on statistical significance. Substantial changes occurred on P value (55.66% increased and 44.12% decreased) and the width of 95% CI (50.68% inflated and 49.32% declined) when excluding studies with no events. Simulation study confirmed these findings.

CONCLUSION

Studies with no events in both arms are not necessarily noninformative. Excluding such studies may alter conclusions.

Environmental enrichment and a selective metabotropic glutamate receptor2/3 (mGluR2/3) agonist suppress amphetamine self-administration: Characterizing baseline differences

A challenge for developing effective treatments for substance use disorders (SUDs) is understanding how environmental variables alter the efficacy of therapeutics. Environmental enrichment (EC) enhances brain development and protects against behaviors associated with drug abuse vulnerability when compared to rats reared in isolation (IC) or standard conditions (SC). EC rearing enhances the expression and function of metabotropic glutamate receptor_2/3 (mGlurR_2/3) and activating mGluR_2/3 reduces psychostimulant self-administration (SA). However, the ability for mGluR_2/3 activation to suppress amphetamine (AMP) SA in differentially reared rats is not determined. Therefore, we tested the hypothesis EC reduces AMP (SA) by augmenting mGluR_2/3 function. At postnatal day 21, male Sprague-Dawley rats were assigned to EC, IC, or SC environments for 30 days. Then, they acquired AMP SA and were moved to a progressive ratio (PR) schedule of reinforcement. EC, IC, and SC rats were pretreated with LY379268 (vehicle, 0.3 and 1 mg/kg), a selective mGluR_2/3 agonist, before PR behavioral sessions. Linear mixed effects analysis determined EC rats had reduced motivation for AMP SA when compared to IC or SC rats and that LY379268 dose-dependently suppressed AMP SA, but there was no evidence of an interaction. Cumming/Gardner-Altman estimation plots illustrate that the 0.3 mg/kg dose suppressed infusions in EC rats while the 1 mg/kg dose suppressed infusions in SC rats. LY379268 was incapable of suppressing the motivation for AMP SA in IC rats. Controlling for baseline differences in differentially reared rats remains a challenge. Normalizing to a baseline introduced error which is illustrated in the precision of the estimated effect size differences. The data indicate that environmental enrichment enhances the ability of a selective mGluR_2/3 agonist to suppress AMP SA and indicates the functional status of the mGluR_2/3 is formed during development. Therefore, environmental history must be considered when evaluating pharmacological therapeutics particularly those aimed at the mGluR_2/3.

A Reckless Guide to P-values : Local Evidence, Global Errors

This chapter demystifies P-values, hypothesis tests and significance tests and introduces the concepts of local evidence and global error rates. The local evidence is embodied in this data and concerns the hypotheses of interest for this experiment, whereas the global error rate is a property of the statistical analysis and sampling procedure. It is shown using simple examples that local evidence and global error rates can be, and should be, considered together when making inferences. Power analysis for experimental design for hypothesis testing is explained, along with the more locally focussed expected P-values. Issues relating to multiple testing, HARKing and P-hacking are explained, and it is shown that, in many situations, their effects on local evidence and global error rates are in conflict, a conflict that can always be overcome by a fresh dataset from replication of key experiments. Statistics is complicated, and so is science. There is no singular right way to do either, and universally acceptable compromises may not exist. Statistics offers a wide array of tools for assisting with scientific inference by calibrating uncertainty, but statistical inference is not a substitute for scientific inference. P-values are useful indices of evidence and deserve their place in the statistical toolbox of basic pharmacologists.

Decentralized distribution-sampled classification models with application to brain imaging

BACKGROUND

In this age of big data, certain models require very large data stores in order to be informative and accurate. In many cases however, the data are stored in separate locations requiring data transfer between local sites which can cause various practical hurdles, such as privacy concerns or heavy network load. This is especially true for medical imaging data, which can be constrained due to the health insurance portability and accountability act (HIPAA) which provides security protocols for medical data. Medical imaging datasets can also contain many thousands or millions of features, requiring heavy network load.

NEW METHOD

Our research expands upon current decentralized classification research by implementing a new singleshot method for both neural networks and support vector machines. Our approach is to estimate the statistical distribution of the data at each local site and pass this information to the other local sites where each site resamples from the individual distributions and trains a model on both locally available data and the resampled data. The model for each local site produces its own accuracy value which are then averaged together to produce the global average accuracy.

RESULTS

We show applications of our approach to handwritten digit classification as well as to multi-subject classification of brain imaging data collected from patients with schizophrenia and healthy controls. Overall, the results showed comparable classification accuracy to the centralized model with lower network load than multishot methods.

COMPARISON WITH EXISTING METHODS

Many decentralized classifiers are multishot, requiring heavy network traffic. Our model attempts to alleviate this load while preserving prediction accuracy.

CONCLUSIONS

We show that our proposed approach performs comparably to a centralized approach while minimizing network traffic compared to multishot methods.

Large-scale directed network inference with multivariate transfer entropy and hierarchical statistical testing

Network inference algorithms are valuable tools for the study of large-scale neuroimaging datasets. Multivariate transfer entropy is well suited for this task, being a model-free measure that captures nonlinear and lagged dependencies between time series to infer a minimal directed network model. Greedy algorithms have been proposed to efficiently deal with high-dimensional datasets while avoiding redundant inferences and capturing synergistic effects. However, multiple statistical comparisons may inflate the false positive rate and are computationally demanding, which limited the size of previous validation studies. The algorithm we present-as implemented in the IDTxl open-source software-addresses these challenges by employing hierarchical statistical tests to control the family-wise error rate and to allow for efficient parallelization. The method was validated on synthetic datasets involving random networks of increasing size (up to 100 nodes), for both linear and nonlinear dynamics. The performance increased with the length of the time series, reaching consistently high precision, recall, and specificity (>98% on average) for 10,000 time samples. Varying the statistical significance threshold showed a more favorable precision-recall trade-off for longer time series. Both the network size and the sample size are one order of magnitude larger than previously demonstrated, showing feasibility for typical EEG and magnetoencephalography experiments.

Principled Missing Data Treatments

We review a number of issues regarding missing data treatments for intervention and prevention researchers. Many of the common missing data practices in prevention research are still, unfortunately, ill-advised (e.g., use of listwise and pairwise deletion, insufficient use of auxiliary variables). Our goal is to promote better practice in the handling of missing data. We review the current state of missing data methodology and recent missing data reporting in prevention research. We describe antiquated, ad hoc missing data treatments and discuss their limitations. We discuss two modern, principled missing data treatments: multiple imputation and full information maximum likelihood, and we offer practical tips on how to best employ these methods in prevention research. The principled missing data treatments that we discuss are couched in terms of how they improve causal and statistical inference in the prevention sciences. Our recommendations are firmly grounded in missing data theory and well-validated statistical principles for handling the missing data issues that are ubiquitous in biosocial and prevention research. We augment our broad survey of missing data analysis with references to more exhaustive resources.