Model selection versus traditional hypothesis testing in circular statistics: a simulation study.
Ontology highlight
ABSTRACT:
SUBMITTER: Landler L
PROVIDER: S-EPMC7327993 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BackgroundFor data collected on a circular rather than linear scale, a very common procedure is to test whether the underlying distribution appears to deviate from circular uniformity. Rao's spacing test is often used to evaluate the support the data offers for the null hypothesis of uniformity. Here we demonstrate that the traditional version of this test fails to adequately control type I error rate when the data is non-continuous (i.e. is rounded/grouped to a finite number of discrete values, e.g. to the nearest degree, a common situation). To overcome this issue, we provide a numerically-intensive simulation version of the test.MethodsWe use a simulation study to explore the performance of the traditional and our novel variant on Rao's spacing test, both in terms of control of type I error rate and statistical power.ResultsWhen data is measured on a continuous circular scale then both methods offer good control of type I error and similar statistical power. If the data is rounded (even to a relatively fine scale such as to the nearest degree - giving 360 possible values), however, the traditional method produces highly inflated type I error rates, particularly with high sample sizes, that make it inappropriate for application to such data. In contrast, our simulation method retains good control of type I error while offering levels of statistical power similar to the traditional Rao test.ConclusionsThe traditional method of applying Rao's spacing test should be replaced by the simulation-based variant introduced here. The two methods offer similar performance but only the simulation method retains good control of the type I error rate when circular data is rounded to a finite set of values (likely due to limited precision of measuring equipment). Adoption of the simulation variant will substantially improve the reliability of this regularly-used test in the commonplace situation where data values are rounded.
| S-EPMC6511169 | biostudies-literature
Project description:BACKGROUND: Repeated exposure to pediatric emergency scenarios improves technical skills, but it is unclear whether repeated exposure to specific cases affects medical decision making in varied cases. OBJECTIVE: We sought to determine whether repeated exposure to 1 scenario would translate to improved performance and decision making in varied scenarios. METHODS: Senior pediatrics residents participated in 3 scenarios with scripted debriefing. Residents were randomized to repeated practice (RP) scenarios or mixed (MIX) scenarios. RP residents completed pulseless electrical activity (PEA) with different stems (Case 1, 2, 3). MIX residents completed PEA (Case 1), seizure (Case 2), and ventricular tachycardia (Case 3) scenarios. Four months later, participants returned to complete 3 more cases: PEA (Case 4), seizure (Case 5), and critical coarctation (Case 6). RESULTS: Twenty-three residents participated in the study and were randomized to either the RP or the MIX group. The RP group showed statistically significant improvement in time to start chest compressions, whereas the MIX group showed no improvement. Use of a backboard improved significantly in Case 4 for the RP group but not for the MIX group. Similarly, time to check glucose in the seizure scenario was significantly better in the MIX group that had previous exposure to a seizure scenario. No differences in performance were noted between groups in Case 6, which was new to both groups. CONCLUSIONS: Results of this study indicate that whereas repeated exposure may improve decision-making skills in similar scenarios, it may not translate to improved medical decision making in other scenarios.
| S-EPMC4054726 | biostudies-other
Project description:Neuroimaging studies typically identify neural activity correlated with the predictions of highly parameterized models, like the many reward prediction error (RPE) models used to study reinforcement learning. Identified brain areas might encode RPEs or, alternatively, only have activity correlated with RPE model predictions. Here, we use an alternate axiomatic approach rooted in economic theory to formally test the entire class of RPE models on neural data. We show that measurements of human neural activity from the striatum, medial prefrontal cortex, amygdala, and posterior cingulate cortex satisfy necessary and sufficient conditions for the entire class of RPE models. However, activity measured from the anterior insula falsifies the axiomatic model, and therefore no RPE model can account for measured activity. Further analysis suggests the anterior insula might instead encode something related to the salience of an outcome. As cognitive neuroscience matures and models proliferate, formal approaches of this kind that assess entire model classes rather than specific model exemplars may take on increased significance.
| S-EPMC2957369 | biostudies-literature
Project description:Hepcidin, the iron-regulatory hormone and acute phase reactant, is proposed to contribute to the pathogenesis of atherosclerosis by promoting iron accumulation in plaque macrophages, leading to increased oxidative stress and inflammation in the plaque (the "iron hypothesis"). Hepcidin and iron may thus represent modifiable risk factors in atherosclerosis. We measured hepcidin expression in Apoe(-/-) mice with varying diets and ages. To assess the role of macrophage iron in atherosclerosis, we generated Apoe(-/-) mice with macrophage-specific iron accumulation by introducing the ferroportin ffe mutation. Macrophage iron loading was also enhanced by intravenous iron injection. Contrary to the iron hypothesis, we found that hepatic hepcidin expression was not increased at any stage of the atherosclerosis progression in Apoe(-/-) or Apoe/ffe mice and that the atherosclerotic plaque size was not increased in mice with elevated macrophage iron. Our results strongly argue against any significant role of macrophage iron in atherosclerosis progression in mice.
| S-EPMC3880128 | biostudies-literature
Project description:The log-rank test has been widely used to test treatment effects under the Cox model for censored time-to-event outcomes, though it may lose power substantially when the model's proportional hazards assumption does not hold. In this article, we consider an extended Cox model that uses B-splines or smoothing splines to model a time-varying treatment effect and propose score test statistics for the treatment effect. Our proposed new tests combine statistical evidence from both the magnitude and the shape of the time-varying hazard ratio function, and thus are omnibus and powerful against various types of alternatives. In addition, the new testing framework is applicable to any choice of spline basis functions, including B-splines, and smoothing splines. Simulation studies confirm that the proposed tests performed well in finite samples and were frequently more powerful than conventional tests alone in many settings. The new methods were applied to the HIVNET 012 Study, a randomized clinical trial to assess the efficacy of single-dose Nevirapine against mother-to-child HIV transmission conducted by the HIV Prevention Trial Network.
| S-EPMC4247822 | biostudies-literature
Project description:Single nucleus ATAC-seq (snATAC-seq) experimental designs have become increasingly complex with multiple factors that might affect chromatin accessibility, including cell type, tissue of origin, sample location, batch, etc., whose compound effects are difficult to test by existing methods. In addition, current snATAC-seq data present statistical difficulties due to their sparsity and variations in individual sequence capture. To address these problems, we present a zero-adjusted statistical model, PACS, that can allow complex hypothesis testing of factors that affect accessibility while accounting for sparse and incomplete data. For differential accessibility analysis, PACS controls the false positive rate and achieves on average a 17% to 122% higher power than existing tools. We demonstrate the effectiveness of PACS through several analysis tasks including supervised cell type annotation, compound hypothesis testing, batch effect correction, and spatiotemporal modeling. We apply PACS to several datasets from a variety of tissues and show its ability to reveal previously undiscovered insights in snATAC-seq data.
| S-EPMC10418058 | biostudies-literature
Project description:Background:With the development of DNA sequencing technology, static omics profiling in microbial communities, such as taxonomic and functional gene composition determination, has become possible. Additionally, the recently proposed in situ growth rate estimation method allows the applicable range of current comparative metagenomics to be extended to dynamic profiling. However, with this method, the applicable target range is presently limited. Furthermore, the characteristics of coverage depth during replication have not been sufficiently investigated. Results:We developed a probabilistic model that mimics coverage depth dynamics. This statistical model explains the bias that occurs in the coverage depth due to DNA replication and errors that arise from coverage depth observation. Although our method requires a complete genome sequence, it involves a stable to low coverage depth (>0.01×). We also evaluated the estimation using real whole-genome sequence datasets and reproduced the growth dynamics observed in previous studies. By utilizing a circular distribution in the model, our method facilitates the quantification of unmeasured coverage depth features, including peakedness, skewness, and degree of density, around the replication origin. When we applied the model to time-series culture samples, the skewness parameter, which indicates the asymmetry, was stable over time; however, the peakedness and degree of density parameters, which indicate the concentration level at the replication origin, changed dynamically. Furthermore, we demonstrated the activity measurement of multiple replication origins in a single chromosome. Conclusions:We devised a novel framework for quantifying coverage depth dynamics. Our study is expected to serve as a basis for replication activity estimation from a broader perspective using the statistical model.
| S-EPMC7104724 | biostudies-literature
Project description:Polygenic scores quantify the genetic risk associated with a given phenotype and are widely used to predict the risk of complex diseases. There has been recent interest in developing methods to construct polygenic risk scores using summary statistic data. We propose a method to construct polygenic risk scores via penalized regression using summary statistic data and publicly available reference data. Our method bears similarity to existing method LassoSum, extending their framework to the Truncated Lasso Penalty (TLP) and the elastic net. We show via simulation and real data application that the TLP improves predictive accuracy as compared to the LASSO while imposing additional sparsity where appropriate. To facilitate model selection in the absence of validation data, we propose methods for estimating model fitting criteria AIC and BIC. These methods approximate the AIC and BIC in the case where we have a polygenic risk score estimated on summary statistic data and no validation data. Additionally, we propose the so-called quasi-correlation metric, which quantifies the predictive accuracy of a polygenic risk score applied to out-of-sample data for which we have only summary statistic information. In total, these methods facilitate estimation and model selection of polygenic risk scores on summary statistic data, and the application of these polygenic risk scores to out-of-sample data for which we have only summary statistic information. We demonstrate the utility of these methods by applying them to GWA studies of lipids, height, and lung cancer.
| S-EPMC7553329 | biostudies-literature
Project description:Biomaterial properties can be tailored for specific applications via systematic and high-throughput screening of biomaterial–cell interactions. However, progress in material development is often hampered by the lack of adequate in vitro testing methods, frequently due to incomplete understanding of involved in vivo mechanisms. In line with drug discovery in pharmacology, a crucial step in assay development for biomaterial screening is the identification of a target to direct the screen against. Herein, the cell type to be used for screening is of essential importance and has to be carefully chosen. So far, few attention has been put on selecting a cell type specifically suitable for in vitro testing of materials for predefined applications. In this manuscript, we describe an approach to define a suitable cell type for screening assays, for which biomaterials for bone regeneration served as example. Using a bioinformatics methodology, different cell lines are compared on three well-characterized model materials. The transcriptional profiles of MG63, iMSC, SV-HFO, hPPCT, hBPCT and SW480 cells are assessed on 3 calcium phosphate-based materials to evaluate their potential application in a screening assay. We show that MG63 is the most suitable cell line to evaluate biomaterials for bone regeneration applications, evidenced by their robust gene expression differences between the 3 model materials. The gene expression differences between the cell lines were assessed based on the overall gene expression profiles and specific subsets of genes and pathways related to osteogenesis and bone homeostasis in response to the 3 materials tested. In the next phase, this cell line will be used to identify a target correlating with in vivo biomaterial performance and henceforth to develop an in vitro screening system.
2018-08-12 | E-MTAB-7057 | biostudies-arrayexpress
Project description:Detecting combinations of alleles that diverged between subpopulations via selection signature statistics can contribute to decipher the phenomenon of epistasis. This research focused on the simulation of genomic data from subpopulations under divergent epistatic selection (ES). We used D'IS2 and FST statistics in pairs of loci to scan the whole-genome. The results showed the ability to identify loci under additive-by-additive ES (ESaa) by reporting large statistical departures between subpopulations with a high level of divergence, while it did not show the same advantage in the other types of ES. Despite this, limitations such as the difficulty to distinguish between the quasi-complete fixation of one locus by ESaa from other events were observed. However, D'IS2 can detect loci under ESaa by defining a minimum boundary for the minor allele frequency on a multiple subpopulation analysis where ES only takes place in one subset. Even so, the major limitation was distinguishing between ES and single-locus selection (SS); therefore, we can conclude that divergent locus can be also a result of ES. The test conditions with D-statistics of both Ohta (1982a, 1982b) and Black and Krafsur (1985) did not provide evidence to differentiate ES in our simulation framework of isolated subpopulations.
| S-EPMC6355851 | biostudies-other