Title: Vaccine Induced Cellular Immunogenicity with Bivariate Modeling
Version: 0.7.3
Date: 2024-02-02
Description: A shiny app for accurate estimation of vaccine induced immunogenicity with bivariate linear modeling. Method is detailed in: Lhomme, Hejblum, Lacabaratz, Wiedemann, Lelievre, Levy, Thiebaut & Richert (2020). Journal of Immunological Methods, 477:112711. <doi:10.1016/j.jim.2019.112711>.
BugReports: https://github.com/sistm/vici/issues
License: GPL-3
Encoding: UTF-8
LazyData: true
Imports: cowplot, DT, ggplot2, grDevices, ggpubr, nlme, shiny, stats, tidyr, utils, numDeriv, stringr, RColorBrewer, scales, shinyWidgets
Suggests: testthat
RoxygenNote: 7.3.0
NeedsCompilation: no
Packaged: 2024-02-02 16:05:11 UTC; boris
Author: Boris Hejblum [cre, aut], Melanie Huchon [aut], Clement Nerestan [aut]
Maintainer: Boris Hejblum <boris.hejblum@u-bordeaux.fr>
Repository: CRAN
Date/Publication: 2024-02-02 16:20:02 UTC

Toy data to upload in the app.

Description

Toy data to upload in the app.

Usage

data(ICS_ex)

Format

A tab-separated .txt file

Examples

if(interactive()){
set.seed(1382019)
nsubj <- 20
ntp <- 3
nstim <- 3
narm <- 3
subj <- rep(rep(rep(1:nsubj, each = ntp), times = nstim), times = narm)
stim <- rep(rep(c("NS", "S1", "S2"), each = nsubj*ntp), times = narm)
tp <- rep(rep(c("D0", "D1", "D3"), times=nsubj*nstim), times = narm)
a <- rep(c("Placebo", "A2", "A3"), each = nsubj*nstim*ntp)
y1 <- round(abs(rnorm(n=nsubj*nstim*ntp*narm,m = 0.03, sd=0.06)) +
  (stim=="S2" & a == "A2" & tp == "D1")*abs(rnorm(n=nsubj*nstim*ntp*narm, m = 0.05, sd=0.01)), 4)
y2 <- round(abs(rnorm(n=nsubj*nstim*ntp*narm,m = 0.03, sd=0.06)) +
  (stim=="S1" & a =="A3" & tp == "D3")*abs(rnorm(n=nsubj*nstim*ntp*narm, m = 0.1, sd=0.02)), 4)
ICS_ex <- cbind.data.frame("Subject" = subj, "StimulationPool" = stim, "TimePoint" = tp,
                           "Arm" = a, "Response1" = y1, "Response2" = y2)
#View(ICS_ex)
write.table(ICS_ex, file="Documents/GitHub/vici/data/ICS_ex.txt", sep="\t",
row.names = FALSE, quote = FALSE)
}

Plotting function for displaying boxplots and associated p-values

Description

Internal function for displaying significance boxplots

Usage

boxplot_VICI(
  data_df,
  pval_2plot,
  response_name,
  input,
  inter = TRUE,
  baseline = NULL,
  fill = FALSE
)

Arguments

data_df

a data.frame.

pval_2plot

a data.frame with the p-values to display.

response_name

a character string indicating the name of the response.

input

internal input from UI.

inter

a logical flag indicating whether we are in the interarm setting or not. Default is TRUE.

baseline

baseline value used in title when inter is FALSE. Default is NULL.

fill

a logical flag indicating if the boxplot is filled Default if FALSE

Value

a ggpubr plot object

Author(s)

Boris Hejblum

Compute_jaclist quantities needed for the Satterthwaite approximation.

Description

Computes vcov of variance parameters (theta, sigma), jacobian of each variance parameter etc.

Usage

compute_jaclist(object, tol = 1e-06)

Arguments

object

a gls object.

tol

a tolerance

Details

This code is adapted from code in compute_auxillary internal function of pbkrtest package.

Value

a list.

Between-Within functions to obtain Denominator degrees of freedom

Description

Between-Within functions to obtain Denominator degrees of freedom

Usage

ddf_BW(object, L)

Compute Full Deviance

Description

Compute Full Deviance

Usage

devfun_gls(varpar, gls_obj)

Arguments

varpar

variance parameters.

gls_obj

a gls object.

Details

This code is adapted from code in devfun_vp internal function of pbkrtest package.

Value

the full deviance, a numerical scalar.

Functions to obtain coefficient, degree of freedom, p-value

Description

This function allows to calculate the different approximations of degrees of freedom and returns the table of results in the app.

Usage

get_coefmat_gls(
  model,
  ddf = c("Satterthwaite", "Kenward-Roger", "Between-Within")
)

Arguments

model

a gls model.

ddf

degrees of freedom approximation.

Value

a matrix containing coefficient, degrees of freedom and p-value

A heatmap function for displaying

Description

Internal function for displaying significance heatmap when multiple conditions are tested

Usage

heatmap_vici(res_2plot, inter = TRUE, baseline = NULL)

Arguments

res_2plot

a data.frame

inter

a logical flag indicating whether we are in the interarm setting or not. Default is TRUE.

Value

a ggplot2 plot object

Author(s)

Boris Hejblum

Plotting function for displaying histograms and associated p-values

Description

Internal function for displaying significance histograms

Usage

histogram_VICI(
  data_df,
  pval_2plot,
  response_name,
  input,
  inter = TRUE,
  baseline = NULL
)

Arguments

data_df

a data.frame.

pval_2plot

a data.frame with the p-values to display.

response_name

a character string indicating the name of the response.

input

internal input from UI.

inter

a logical flag indicating whether we are in the interarm setting or not. Default is TRUE.

baseline

baseline value used in title when inter is FALSE. Default is NULL.

Value

a ggpubr plot object

Author(s)

Clément NERESTAN

Fitting GLS For Inter-Arm Setting

Description

Fitting GLS For Inter-Arm Setting

Usage

interarm_fit(transformed_data, input, resp)

Fitting GLS For Intra-Arm Setting

Description

Fitting GLS For Intra-Arm Setting

Usage

intraarm_fit(transformed_data, tested_time, input, resp)

mod_modelfit_ui and mod_modelfit_server

Description

A shiny Module.

Usage

mod_modelfit_ui(id)

mod_modelfit_server(input, output, session, datas, parent, origin)

Arguments

id

shiny id

input

internal

output

internal

session

internal

datas

internal

parent

internal

origin

internal

mod_settings_pan_ui and mod_settings_pan_server

Description

A shiny Module.

Usage

mod_settings_pan_ui(id)

mod_settings_pan_server(input, output, session, datas, parent)

Arguments

id

shiny id

input

internal

output

internal

session

internal

datas

internal

parent

Custom download handler for plots

Description

Custom download handler for plots

Usage

myDownloadHandlerForPlots(name, plot_obj, outputArgs = list())

Arguments

name

output file name

plot_obj

a plot object to be downloaded

Value

a ggpubr plot object

Author(s)

Boris Hejblum

Our generalized least squares ls function

Description

Internal function to adapt generalized least squares (gls) model with more details in output.

Usage

mygls(
  model,
  data = sys.frame(sys.parent()),
  correlation = NULL,
  weights = NULL,
  subset,
  method = c("REML", "ML"),
  na.action = na.fail,
  control = list(),
  verbose = FALSE
)

Arguments

model

a formula.

data

a data.frame containing variables called in model, correlation, weights, subset.

correlation

a corCompSymm object. Default is NULL.

weights

a varIdent object. Default is NULL.

subset

an optional expression indicating which subset of the rows of data should be used in the fit. By default, all observations are included.

method

a character string to choose the maximization method. Default is "REML".

na.action

a function that indicates what should happen when the data contain NAs. Default is na.fail.

control

a list of control values. Default is an empty list.

verbose

an optional logical value. If TRUE information on the evolution of the iterative algorithm is printed. Default is FALSE.

Value

a gls object

Compute Quadratic Form

Description

Compute Quadratic Form

Usage

qform(L, V)

Arguments

L

a numeric vector.

V

a symmetric numeric matrix.

Value

a numerical scalar.

rbind Multiple Objects

Description

rbind Multiple Objects

Usage

rbindall(...)

Arguments

...

objects to be rbind'ed - typically matrices or vectors

Launch VICI Shiny App

Description

Launch VICI Shiny App

Usage

run_app(host = "127.0.0.1", port = 3838, ...)

Arguments

host

Default is "127.0.0.1", see runApp for details.

port

Default is 3838, see runApp for details.

...

additional arguments to be passed to the runApp function.

Examples

if(interactive()){
vici::run_app()
}

Compute covariance of Beta for a Generalized Least Squares (GLS) Model

Description

Compute covariance of Beta for a Generalized Least Squares (GLS) Model

Usage

varBetafun_gls(varpar, gls_obj)

Arguments

varpar

variance parameters.

gls_obj

a gls object.

Details

This code is adapted from code in get_covbeta internal function of pbkrtest package.

Value

covariance of Beta, a numerical scalar.

Compute Wald Confidence Interval

Description

Compute Wald Confidence Interval

Usage

waldCI(estimate, se, df = Inf, level = 0.95)

Arguments

estimate

an estimated coefficient.

se

standard error of estimate.

df

degrees of freedom associate to estimate. df = Inf is allowed.

level

level of confidence interval.

Details

This code is greatly inspired by code from the lmerTest package.

Value

a matrix of lower and upper confidence interval.