<b>qgam</b>: Bayesian Nonparametric Quantile Regression Modeling in <i>R</i>

Fasiolo, Matteo; Wood, Simon N.; Zaffran, Margaux; Nedellec, Raphaël; Goude, Yannig

doi:10.18637/jss.v100.i09

Cited by 51 publications

(38 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We take advantage of our empirical observation that sidebands (as in Figure S2) on our 15 T FT-ICR-MS (Bruker Solarix XR) generally deviate from the expected trend of reduced resolution with increasing m/z (Figure S3). We use a median regression 29 where we take m/z as explanatory variable and resolution as response. The density gradient of residuals should always be negative when moving away from the density maximum.…”

Section: ■ Removal Of Sidebands and Contaminantsmentioning

confidence: 99%

“…14 We use nonparametric smoothing models 27 to predict and eliminate such systematic errors (Figure S4). For cases with strong variance increase along the m/z axis, we supply a median regression 29 to cope with such heteroscedasticity. For Orbitrap users we additionally offer a weighted tensor product smoothing supplied by the "mgcv" package, 27 including S/MDL and m/z simultaneously in the model fit.…”

Section: ■ Removal Of Sidebands and Contaminantsmentioning

confidence: 99%

See 1 more Smart Citation

ICBM-OCEAN: Processing Ultrahigh-Resolution Mass Spectrometry Data of Complex Molecular Mixtures

et al. 2020

View full text Add to dashboard Cite

Untargeted molecular analyses of complex mixtures are relevant for many fields of research, including geochemistry, pharmacology, and medicine. Ultrahigh-resolution mass spectrometry is one of the most powerful tools in this context. The availability of open scripts and online tools for specific data processing steps such as noise removal or molecular formula assignment is growing, but an integrative tool where all crucial steps are reproducibly evaluated and documented is lacking. We developed a novel, server-based tool (ICBM-OCEAN, Institute for Chemistry and Biology of the Marine Environment, Oldenburg−complex molecular mixtures, evaluation & analysis) that integrates published and novel approaches for standardized processing of ultrahigh-resolution mass spectrometry data of complex molecular mixtures. Different from published approaches, we offer diagnostic and validation tools for all relevant steps. Among other features, we included objective and reproducible reduction of noise and systematic errors, spectra recalibration and alignment, and identification of likeliest molecular formulas. With 15 chemical elements, the tool offers high flexibility in formula attribution. Alignment of mass spectra among different samples prior to molecular formula assignment improves mass error and facilitates molecular formula confirmation with the help of isotopologues. The online tool and the detailed instruction manual are freely accessible at www.icbm.de/icbm-ocean.

show abstract

Section: ■ Removal Of Sidebands and Contaminantsmentioning

confidence: 99%

Section: ■ Removal Of Sidebands and Contaminantsmentioning

confidence: 99%

ICBM-OCEAN: Processing Ultrahigh-Resolution Mass Spectrometry Data of Complex Molecular Mixtures

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Alternatively, other papers (e.g., Corradin et al 2021;Hosszejni and Kastner 2021;Knaus et al 2021;Venturini and Piccarreta 2021) write sampling functions in C++ which are then integrated into R by using the Rcpp (Eddelbuettel and François 2011) and RcppArmadillo (Eddelbuettel and Sanderson 2014) packages. Finally, some papers do not use MCMC but numerical approximations such as Eggleston et al (2021), Fasiolo et al (2021 and Van Niekerk et al (2021). Two papers (Kuschnig and Vashold 2021;Weber et al 2021) also implement priors for specific Bayesian models.…”

Section: Discussionmentioning

confidence: 99%

Software for Bayesian Statistics

Cameletti

Gómez‐Rubio

2021

J. Stat. Soft.

View full text Add to dashboard Cite

In this summary we introduce the papers published in the special issue on Bayesian statistics. This special issue comprises 20 papers on Bayesian statistics and Bayesian inference on different topics such as general packages for hierarchical linear model fitting, survival models, clinical trials, missing values, time series, hypothesis testing, priors, approximate Bayesian computation, and others.

show abstract

“…For the so‐generated data, we fitted the correctly specified linear mixed‐effects and fixed‐effects models from scenario B (Table 1 , Equations M8 and M10) and calculated type I error rate and statistical power of the population‐level effect. We then fitted a quantile regression using the qgam R‐package (Fasiolo et al, 2020 ), with the statistical property (power and type I error rate) as response and variance, number of levels, total number of observations and the unbalance proxy as splines. We used a quantile regression with splines as we expect a non‐linear relationship.…”

Section: Methodsmentioning

confidence: 99%

Fixed or random? On the reliability of mixed‐effects models for a small number of levels in grouping variables

Oberpriller

Leite

Pichler

2022

Ecology and Evolution

View full text Add to dashboard Cite

Biological data are often intrinsically hierarchical (e.g., species from different genera, plants within different mountain regions), which made mixed‐effects models a common analysis tool in ecology and evolution because they can account for the non‐independence. Many questions around their practical applications are solved but one is still debated: Should we treat a grouping variable with a low number of levels as a random or fixed effect? In such situations, the variance estimate of the random effect can be imprecise, but it is unknown if this affects statistical power and type I error rates of the fixed effects of interest. Here, we analyzed the consequences of treating a grouping variable with 2–8 levels as fixed or random effect in correctly specified and alternative models (under‐ or overparametrized models). We calculated type I error rates and statistical power for all‐model specifications and quantified the influences of study design on these quantities. We found no influence of model choice on type I error rate and power on the population‐level effect (slope) for random intercept‐only models. However, with varying intercepts and slopes in the data‐generating process, using a random slope and intercept model, and switching to a fixed‐effects model, in case of a singular fit, avoids overconfidence in the results. Additionally, the number and difference between levels strongly influences power and type I error. We conclude that inferring the correct random‐effect structure is of great importance to obtain correct type I error rates. We encourage to start with a mixed‐effects model independent of the number of levels in the grouping variable and switch to a fixed‐effects model only in case of a singular fit. With these recommendations, we allow for more informative choices about study design and data analysis and make ecological inference with mixed‐effects models more robust for small number of levels.

show abstract

qgam: Bayesian Nonparametric Quantile Regression Modeling in R

Cited by 51 publications

References 22 publications

ICBM-OCEAN: Processing Ultrahigh-Resolution Mass Spectrometry Data of Complex Molecular Mixtures

ICBM-OCEAN: Processing Ultrahigh-Resolution Mass Spectrometry Data of Complex Molecular Mixtures

Software for Bayesian Statistics

Fixed or random? On the reliability of mixed‐effects models for a small number of levels in grouping variables

Contact Info

Product

Resources

About