Negative Binomial Generalized Linear Model Research Articles

Fungal Metabarcoding Data for Two Grapevine Varieties (Regent and Vitis vinifera ‘Cabernet-Sauvignon’) Inoculated with Powdery Mildew (Erysiphe necator) Under Drought Conditions

Fungal Metabarcoding Data for Two Grapevine Varieties (Regent and <i>Vitis vinifera</i> ‘Cabernet-Sauvignon’) Inoculated with Powdery Mildew (<i>Erysiphe necator</i>) Under Drought Conditions

REAL-WORLD EVIDENCE OF STAGE ASSOCIATED ENDOMETRIOSIS-INDUCED OVARIAN FAILURE

Evaluation of real-world evidence for endometriosis-induced ovarian failure. Patient diagnoses and first cycle AMH, BAFC, and age were extracted from electronic medical records from patients undergoing ovulatory induction with or without IUI and autologous IVF cycles at 11 fertility centers in the Unites States from 2002-2019. The available cohort for each analysis was classified by stage as Endo1, Endo2, Endo3, Endo4, EndoUnspecified, NonEndo (AMH: N=210, 390, 373, 354, 56322; BAFC: N=570, 1117, 866, 901, 869, 107262; Oocytes retrieved: N=102, 181, 182, 306, 162, 18536). The outcomes AMH and BAFC were modeled using a Gaussian family generalized linear model (GLM) with covariates age and endometriosis stage; oocytes retrieved was modeled using a negative binomial GLM with age, BAFC, AMH, and endometriosis stage as covariates. GLM models for AMH and BAFC found that endometriosis is significantly associated with decreased levels of ovarian reserve markers after controlling for age in a stage-dependent manner (p < 0.0001):Tabled 1DiagnosisPercent Decrease Relative to nonEndo (95% Confidence Interval)BAFCAMHEndo1-16.1% (-21.1, -10.6)-18.4% (-31.4, -2.9)Endo2-19.8% (-23.5, -15.9)-25.2% (-35.1, -13.7)Endo3-25.2% (-29.4, -20.8)-37.3% (-47.5, -25.2)Endo4-33.6% (-37.7, -29.4)-42.9% (-51.8, -32.3)EndoUnspecified-20.2% (-24.3, -15.9)-39.6% (-49.3, -28.1) Open table in a new tab Endometriosis has been associated with decreased response to ovarian stimulation and poorer IVF outcomes, however it has not been evaluated whether this is primarily a consequence of endometriosis-induced impact on ovarian reserve. We found that when controlling for age, patient diagnosis of Endo3, Endo4, or EndoUnspecified were found to be significantly associated with decreased oocytes retrieved. Furthermore, after controlling for ovarian reserve metrics AMH and BAFC, the association of endometriosis with oocyte retrieved is no longer significant. The detrimental effects of endometriosis on ovarian reserve are associated with severity of disease, with the impact on ovarian quantity being a primary driver of reduced response to ovarian stimulation observed in patients with endometriosis.

An Efficient and Flexible Spike Train Model Via Empirical Bayes

Accurate statistical models of neural spike responses can characterize the information carried by neural populations. But the limited samples of spike counts during recording usually result in model overfitting. Besides, current models assume spike counts to be Poisson-distributed, which ignores the fact that many neurons demonstrate over-dispersed spiking behaviour. Although the Negative Binomial Generalized Linear Model (NB-GLM) provides a powerful tool for modeling over-dispersed spike counts, the maximum likelihood-based standard NB-GLM leads to highly variable and inaccurate parameter estimates. Thus, we propose a hierarchical parametric empirical Bayes method to estimate the neural spike responses among neuronal population. Our method integrates both Generalized Linear Models (GLMs) and empirical Bayes theory, which aims to (1) improve the accuracy and reliability of parameter estimation, compared to the maximum likelihood-based method for NB-GLM and Poisson-GLM; (2) effectively capture the over-dispersion nature of spike counts from both simulated data and experimental data; and (3) provide insight into both neural interactions and spiking behaviours of the neuronal populations. We apply our approach to study both simulated data and experimental neural data. The estimation of simulation data indicates that the new framework can accurately predict mean spike counts simulated from different models and recover the connectivity weights among neural populations. The estimation based on retinal neurons demonstrate the proposed method outperforms both NB-GLM and Poisson-GLM in terms of the predictive log-likelihood of held-out data. <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup>

Mapping wind erosion hazard with regression-based machine learning algorithms

Land susceptibility to wind erosion hazard in Isfahan province, Iran, was mapped by testing 16 advanced regression-based machine learning methods: Robust linear regression (RLR), Cforest, Non-convex penalized quantile regression (NCPQR), Neural network with feature extraction (NNFE), Monotone multi-layer perception neural network (MMLPNN), Ridge regression (RR), Boosting generalized linear model (BGLM), Negative binomial generalized linear model (NBGLM), Boosting generalized additive model (BGAM), Spline generalized additive model (SGAM), Spike and slab regression (SSR), Stochastic gradient boosting (SGB), support vector machine (SVM), Relevance vector machine (RVM) and the Cubist and Adaptive network-based fuzzy inference system (ANFIS). Thirteen factors controlling wind erosion were mapped, and multicollinearity among these factors was quantified using the tolerance coefficient (TC) and variance inflation factor (VIF). Model performance was assessed by RMSE, MAE, MBE, and a Taylor diagram using both training and validation datasets. The result showed that five models (MMLPNN, SGAM, Cforest, BGAM and SGB) are capable of delivering a high prediction accuracy for land susceptibility to wind erosion hazard. DEM, precipitation, and vegetation (NDVI) are the most critical factors controlling wind erosion in the study area. Overall, regression-based machine learning models are efficient techniques for mapping land susceptibility to wind erosion hazards.

Impact of hydrological factors on the dynamic of COVID-19 epidemic: A multi-region study in China

Considering the live SARS-CoV-2 was detected and isolated from the excrement and urine of infected patients, the potential public health risk of its waterborne transmission should be paid broad and close attention. The purpose of the current study is to investigate the associations between COVID-19 incidences and hydrological factors such as lake area, river length, precipitation and volume of water resources in 30 regions of China. All confirmed cases for each areas were divided into two clusters including first cases cluster driven by imported cases during the period of January 20th to January 29th, 2020 and second cases cluster driven by local cases during the period of January 30th to March 1st, 2020. Based on the results of descriptive analysis and nonlinear regression analysis, positive associations with COVID-19 confirmed numbers were observed for migration scale index (MSI), river length, precipitation and volume of water resources, but negative associations for population density. The correlation coefficient in the second stage cases cluster is apparently higher than that in the first stage cases cluster. Then, the negative binomial-generalized linear model (NB-GLM) was fitted to estimate area-specific effects of hydrological variables on relative risk (RR) with the incorporation of additional variables (e.g., MSI) and the effects of exposure-lag-response. The statistically significant associations between RR and river length, the volume of water resources, precipitation were obtained by meta-analysis as 1.24 (95% CI: 1.22, 1.27), 2.56 (95% CI: 2.50, 2.61) and 1.59 (95% CI: 1.56, 1.62), respectively. The possible water transmission routes of SARS-CoV-2 and the potential capacity of long-distance transmission of SARS-CoV-2 in water environment was also discussed. Our results could provide a better guidance for local and global authorities to broaden the mind for understanding the natural-social system or intervening measures for COVID-19 control at the current or futural stage.

Effect of Silver Diamine Fluoride Treatment on Microbial Profiles of Plaque Biofilms from Root/Cervical Caries Lesions

Purpose: To assess the effect of silver diamine fluoride (SDF) on microbial profiles present in plaque from root/cervical carious lesions, and its association with caries lesion arrest. Materials and Methods: Twenty patients with at least one soft cavitated root/cervical carious lesion were included. One lesion/patient was randomly selected and treated with 38% SDF. Supragingival plaque samples were harvested at preintervention and 1 month postintervention. Using an MiSeq platform, 16S rDNA sequencing of the V3-V4 regions was used to determine bacterial profiles. Clinical evaluation of lesion hardness was used to evaluate arrest. t tests, principal component analysis (PCA), multidimensional scaling (MDS), and generalized linear models (GLMs) tests were used for statistical comparisons. Results: From a total of 40 plaque samples, 468 probe targets were observed. Although 60% of lesions became hard postintervention, PCA and MDS tests showed no distinct pre- and postintervention groups. In addition, pre- and postintervention differences in diversity (Shannon index) of microbial profiles between patients with and without lesion arrest were not statistically different. A likelihood ratio test for pre- versus postintervention differences within patients, i.e., adjusting for differences between patients using negative binomial GLMs, showed 17 bacterial taxa with significant differences (FDR <0.05). Conclusion: Although 60% of lesions hardened after SDF treatment, this was not directly due to either overall statistically significant differences in microbial profiles or differences in microbial diversity. Nevertheless, there was a trend with some acid-producing species in that their relative abundance was reduced postintervention. The negative binomial GLMs showed 17 bacterial taxa that were significantly different after SDF treatment.

Multivariate count data generalized linear models: Three approaches based on the Sarmanov distribution

Starting from the question: What is the accident risk of an insured individual?, we consider that the customer has contracted policies in different insurance lines: motor and home. Three models based on the multivariate Sarmanov distribution are analyzed. Driven by a real data set that takes into account three types of accident risks, two for motor and one for home, three trivariate Sarmanov distributions with generalized linear models (GLMs) for marginals are considered and fitted to the data. To estimate the parameters of these three models, we discuss a method for approaching the maximum likelihood (ML) estimators. Finally, the three models are compared numerically with the simpler trivariate Negative Binomial GLM and with elliptical copula based models.

Multivariate Count Data Generalized Linear Models: Three Approaches Based on the Sarmanov Distribution

Starting from the question: “What is the accident risk of an insured?”, this paper considers a multivariate approach by taking into account three types of accident risks and the possible dependence between them. Driven by a real data set, we propose three trivariate Sarmanov distributions with generalized linear models (GLMs) for marginals and incorporate various individual characteristics of the policyholders by means of explanatory variables. Since the data set was collected over a longer time period (10 years), we also added each individual’s exposure to risk. To estimate the parameters of the three Sarmanov distributions, we analyze a pseudo-maximumlikelihood method. Finally, the three models are compared numerically with the simpler trivariate Negative Binomial GLM.

Statistical models for the number of successful cyber intrusions

We propose several generalized linear models (GLMs) to predict the number of successful cyber intrusions (or “intrusions”) into an organization’s computer network, where the rate at which intrusions occur is a function of the following observable characteristics of the organization: (i) domain name system (DNS) traffic classified by their top-level domains (TLDs); (ii) the number of network security policy violations; and (iii) a set of predictors that we collectively call the “cyber footprint” that is comprised of the number of hosts on the organization’s network, the organization’s similarity to educational institution behavior, and its number of records on scholar.google.com . In addition, we evaluate the number of intrusions to determine whether these events follow a Poisson or negative binomial (NB) probability distribution. We reveal that the NB GLM provides the best fit model for the observed count data, number of intrusions per organization, because the NB model allows the variance of the count data to exceed the mean. We also show that there are restricted and simpler NB regression models that omit selected predictors and improve the goodness-of-fit of the NB GLM for the observed data. With our model simulations, we identify certain TLDs in the DNS traffic as having a significant impact on the number of intrusions. In addition, we use the models and regression results to conclude that the number of network security policy violations is consistently predictive of the number of intrusions.

Ecotoxicology is not normal

Ecotoxicologists often encounter count and proportion data that are rarely normally distributed. To meet the assumptions of the linear model, such data are usually transformed or non-parametric methods are used if the transformed data still violate the assumptions. Generalized linear models (GLMs) allow to directly model such data, without the need for transformation. Here, we compare the performance of two parametric methods, i.e., (1) the linear model (assuming normality of transformed data), (2) GLMs (assuming a Poisson, negative binomial, or binomially distributed response), and (3) non-parametric methods. We simulated typical data mimicking low replicated ecotoxicological experiments of two common data types (counts and proportions from counts). We compared the performance of the different methods in terms of statistical power and Type I error for detecting a general treatment effect and determining the lowest observed effect concentration (LOEC). In addition, we outlined differences on a real-world mesocosm data set. For count data, we found that the quasi-Poisson model yielded the highest power. The negative binomial GLM resulted in increased Type I errors, which could be fixed using the parametric bootstrap. For proportions, binomial GLMs performed better than the linear model, except to determine LOEC at extremely low sample sizes. The compared non-parametric methods had generally lower power. We recommend that counts in one-factorial experiments should be analyzed using quasi-Poisson models and proportions from counts by binomial GLMs. These methods should become standard in ecotoxicology.

Moving to the sea: a challenge for an inshore species, the slender-billed gull

MEPS Marine Ecology Progress Series Contact the journal Facebook Twitter RSS Mailing List Subscribe to our mailing list via Mailchimp HomeLatest VolumeAbout the JournalEditorsTheme Sections MEPS 463:285-295 (2012) - DOI: https://doi.org/10.3354/meps09861 Moving to the sea: a challenge for an inshore species, the slender-billed gull Albert Cama1,2,5,*, Rosa Abellana3, Isadora Christel1,2, Xavier Ferrer1, David R. Vieites2,4 1Institut de Recerca de la Biodiversitat i Departament de Biologia Animal, Facultat de Biologia, Universitat de Barcelona, 08028 Barcelona, Spain 2Museo Nacional de Ciencias Naturales (MNCN), Consejo Superior de Investigaciones Científicas (CSIC), 28006 Madrid, Spain 3Departament de Salut Pública, Facultat de Medicina, Universitat de Barcelona, 08036 Barcelona, Spain 4REFER Biodiversity Chair, University of Porto, CIBIO, Campus Agrário de Vairão, R. Padre Armando Quintas, 4485-661 Vairão, Portugal 5Present address: SEO/BirdLife. C/Murcia 2-8, Local 13, 08026 Barcelona, Spain *Email: albert_camatorrell@yahoo.es ABSTRACT: The slender-billed gull Chroicocephalus genei is a typical saltpan and brackish lagoon inshore forager. In 2005, there were two breeding colonies in the Ebro Delta (western Mediterranean): one south of the delta on a large saltpan, and the other one in the north, where typical foraging habitats are scarce. Aerial surveys showed that gulls foraged at sea throughout the study period, but were mainly restricted to the northern part of the delta from June to August. We compared 6 different approaches to model the gulls’ distribution at sea: Poisson generalized linear model (GLM), zero inflated Poisson GLM, spatial Poisson generalized linear mixed model (GLMM), non-spatial Poisson GLMM, negative binomial GLM and zero inflated negative binomial GLM. The best approach, simple negative binomial GLM, suggested that the gulls’ distribution was mainly influenced by the proximity to the colony, tern density and water temperature. The water temperature was indicative of river runoff, which resulted in an anchovy population explosion, explaining the gulls’ extensive foraging at sea in summer. Slender-billed gulls are capable of changing their foraging strategy to exploit resources at sea when occupying an area with limited terrestrial but abundant marine resources. Population growth in the western Mediterranean may lead to the occupancy of non-typical habitats elsewhere. Because slender-billed gulls are rare and have an unfavorable conservation status, conservation policies affecting this species should be extended to the sea. KEY WORDS: Chroicocephalus genei · Distribution modeling · Ebro Delta · Mediterranean Sea · Anchovy · Generalized linear model (GLM) · Generalized linear mixed model (GLMM) Full text in pdf format PreviousNextCite this article as: Cama A, Abellana R, Christel I, Ferrer X, Vieites DR (2012) Moving to the sea: a challenge for an inshore species, the slender-billed gull. Mar Ecol Prog Ser 463:285-295. https://doi.org/10.3354/meps09861 Export citation RSS - Facebook - Tweet - linkedIn Cited by Published in MEPS Vol. 463. Online publication date: August 30, 2012 Print ISSN: 0171-8630; Online ISSN: 1616-1599 Copyright © 2012 Inter-Research.