| Title: | IN-vivo reSPonsE Classification of Tumours |
| Version: | 0.1.0 |
| Maintainer: | Bairu Zhang <bairu.zhang@astrazeneca.com> |
| Description: | This is a shiny app used for the statistical classifying and analysing pre-clinical tumour responses. |
| License: | Apache License (== 2) |
| Depends: | R (≥ 3.5.0) |
| Imports: | brms, dplyr, DT, ggeffects, ggplot2, knitr, lme4, modelr, pander, plotly, purrr, readxl, rlang, rmarkdown, shiny, shinyalert, shinyFeedback, shinyjs, shinytoastr, shinyvalidate, tidybayes, tippy, tidyr, vroom, waiter |
| Suggests: | spelling, lintr, testthat |
| Encoding: | UTF-8 |
| LazyData: | true |
| RoxygenNote: | 7.1.2 |
| Language: | en-US |
| NeedsCompilation: | no |
| Packaged: | 2022-05-05 09:35:58 UTC; kjrg087 |
| Author: | Bairu Zhang [cre, aut], Olga Muraeva [aut], Natasha Karp [aut] |
| Repository: | CRAN |
| Date/Publication: | 2022-05-06 12:10:02 UTC |
create a table with aggregated data: each row contains information about control and treatments of a single study
Description
create a table with aggregated data: each row contains information about control and treatments of a single study
Usage
aggregate_study_info(df)
Arguments
df |
data.frame |
Value
data.frame
Generate table representing number of animals in classification groups
Description
Generate table representing number of animals in classification groups
Usage
animal_info_classification(data)
Arguments
data |
final classification data |
Value
data frame
Credible interval (or say “Bayesian confidence interval”) of the mean difference between two groups (treatment and reference) is used to assess the efficacy. If 0 falls outside the interval, the drug was considered significantly effective
Description
Credible interval (or say “Bayesian confidence interval”) of the mean difference between two groups (treatment and reference) is used to assess the efficacy. If 0 falls outside the interval, the drug was considered significantly effective
Usage
assess_efficacy(data, reference = "Control")
Arguments
data |
prediction results |
reference |
name of the reference treatment |
Value
dataframe with information about drug efficacy
makes df with data to be excluded
Description
makes df with data to be excluded
Usage
below_min_points(df, min_points)
Arguments
df |
initial data frame |
min_points |
minimum number of data points for one animal_id per study |
Value
df
Function to return rate of growth (e.g. the slope after a log transformation of the tumour data against time)
Description
Function to return rate of growth (e.g. the slope after a log transformation of the tumour data against time)
Usage
calc_gr(df, log_tv = "log_tv", day = "day")
Arguments
df |
subset, one animal_id |
log_tv |
name of the column, tumour volume |
day |
name of the column, days |
Value
tibble with GR and GR_SE
Calculate probability of categories
Description
Calculate probability of categories
Usage
calc_probability(data)
Arguments
data |
data frame with predictions |
Value
data frame
Calculate percentage of survived animals
Description
Calculate percentage of survived animals
Usage
calc_survived(df)
Arguments
df |
data frame |
Value
data frame
Get an array with change_time for studies from the population-level effects, multiple studies
Description
Get an array with change_time for studies from the population-level effects, multiple studies
Usage
change_time_multi(model)
Arguments
model |
an object of class brmsfit |
Value
data frame
Get a change time from the population-level effects, single study
Description
Get a change time from the population-level effects, single study
Usage
change_time_single(model)
Arguments
model |
an object of class brmsfit |
Value
a numeric vector of length one
Classify individual data points as Responders or Non-responders
Description
Classify individual data points as Responders or Non-responders
Usage
classify_data_point(df_newstudy, pred_newstudy)
Arguments
df_newstudy |
data from new study |
pred_newstudy |
data frame with predictions |
Value
data frame with "Responder"/"Non-responder" for individual data points
Make predictions for subcategories
Description
Make predictions for subcategories
Usage
classify_subcategories(data, model)
Arguments
data |
data frame with classification results |
model |
object of class brmsfit |
Value
data frame
Classify tumour based on the growth rate and the p_value for a two-sided T test Tumour will be considered as "Non-responder", "Modest responder", "Stable responder" or "Regressing responder"
Description
Classify tumour based on the growth rate and the p_value for a two-sided T test Tumour will be considered as "Non-responder", "Modest responder", "Stable responder" or "Regressing responder"
Usage
classify_type_responder(df)
Arguments
df |
data frame |
Value
data frame with a new column classify_tumour
function to remove hyphens, underscores, spaces and transform to lowercase
Description
function to remove hyphens, underscores, spaces and transform to lowercase
Usage
clean_string(string)
Arguments
string |
to modify |
Value
modified string
Function to plot a control growth profile
Description
Function to plot a control growth profile
Usage
control_growth_plot(df, model_type, col_palette)
Arguments
df |
data frame |
model_type |
string |
col_palette |
character palette |
Value
ggplot object
Tumour volume data over time for in-vivo studies
Description
A dataset containing the repeatedly measurements of tumour volume data over time for individual animals.
Usage
example_data
Format
A data frame with 1462 rows and 6 variables:
- study
study identifier
- group
group identifier
- treatment
treatment type
- animal_id
animal identifier
- day
day after implant
- tumour_volume
volume in mm3
Filter rows to exclude from the analysis
Description
Filter rows to exclude from the analysis
Usage
exclude_data(df, study_id_ex, animal_id_ex, day_ex, reason)
Arguments
df |
initial df |
study_id_ex |
string: study id |
animal_id_ex |
string: animal id |
day_ex |
string: day |
reason |
string: why it should be excluded |
Value
dataframe with rows that meets exclusion criteria
Function to expand a vector of colors if needed
Description
Function to expand a vector of colors if needed
Usage
expand_palette(col_palette, n)
Arguments
col_palette |
character palette to color the treatments |
n |
how many colors are needed |
Value
a character vector of colors
Calculate coefficients for a nonlinear model
Description
Calculate coefficients for a nonlinear model
Usage
f_start(df, x, y, r_change)
Arguments
df |
data frame with x as a predictor and y is an outcome |
x |
predictor string |
y |
outcome string |
r_change |
numeric |
Value
list of coefficients
Classify tumour based on response status of individuals
Description
Classify tumour based on response status of individuals
Usage
get_responder(x, n)
Arguments
x |
character vector with response statuses of one animal |
n |
consecutive measurements for classification |
Value
"Responder" or "Non-responder"
function to search for the possible critical columns in a data.frame
Description
function to search for the possible critical columns in a data.frame
Usage
guess_match(colnames_df, crit_cols)
Arguments
colnames_df |
a character vector with names |
crit_cols |
a character vector |
Value
list: possible match to each critical column
Function to hide outliers in boxplots with jitterdodge as suggested
Description
Function to hide outliers in boxplots with jitterdodge as suggested
Usage
hide_outliers(x)
Arguments
x |
plotly object |
Value
plotly object without boxplot outliers
function to read data from users (.csv or .xlsx files)
Description
function to read data from users (.csv or .xlsx files)
Usage
load_data(path, name)
Arguments
path |
path to a temp file |
name |
filename provided by the web browser |
Value
data frame
Create a character vector with the names of terms from model, for which predictions should be displayed Specific values are specified in square brackets
Description
Create a character vector with the names of terms from model, for which predictions should be displayed Specific values are specified in square brackets
Usage
make_terms(days, studies = NULL)
Arguments
days |
vector with days with which to predict |
studies |
vector with studies with which to predict |
Value
vector with values for predictions
Build model and make predictions
Description
Build model and make predictions
Usage
model_control(df_control, df_newstudy, method, end_day)
Arguments
df_control |
data frame with control data (including historical control, if provided) |
df_newstudy |
data frame, data from new study |
method |
"Two-stage non-linear model" or "Linear model" |
end_day |
period of time used for the statistical modelling of the control data |
Value
list: two data frames with prediction results (for new study and for control data)
Display a popup message and reset fileInput
Description
Display a popup message and reset fileInput
Usage
notify_error_and_reset_input(message_text)
Arguments
message_text |
the modal's text |
Fit model (Bayesian ordered logistic regression)
Description
Fit model (Bayesian ordered logistic regression)
Usage
ordered_regression(df, formula, n_cores)
Arguments
df |
data frame with classification results. Tumour classification is converted into ordinal data |
formula |
string |
n_cores |
number of cores to use |
Value
object of class brmsfit
Plot representing number of animals in classification groups
Description
Plot representing number of animals in classification groups
Usage
plot_animal_info(data, col_palette)
Arguments
data |
final classification data |
col_palette |
character palette |
Value
ggplot object
Function to plot classification over growth rate
Description
Function to plot classification over growth rate
Usage
plot_class_gr(df, col_palette)
Arguments
df |
data frame |
col_palette |
character palette |
Value
ggplot object
Function to plot classification over tumour volume
Description
Function to plot classification over tumour volume
Usage
plot_class_tv(df, col_palette, title_name)
Arguments
df |
data frame |
col_palette |
named vector |
title_name |
character |
Value
ggplot object
Plot estimated proportions
Description
Plot estimated proportions
Usage
plot_proportions(data, col_palette)
Arguments
data |
table of the category prediction |
col_palette |
character palette |
Function to plot waterfall
Description
Function to plot waterfall
Usage
plot_waterfall(df, col_palette, study_name)
Arguments
df |
data frame |
col_palette |
character palette |
study_name |
string: to show on title |
Value
ggplot object
Create volume plot for one-batch data
Description
Create volume plot for one-batch data
Usage
plotly_volume(
df,
col_palette = NULL,
faceting_var,
y_name,
y_var,
p_title,
...
)
Arguments
df |
data.frame, single-batch long format |
col_palette |
character palette to color the treatments |
faceting_var |
string |
y_name |
string |
y_var |
string: column name for y axis |
p_title |
plot title |
... |
arguments passed to plot_ly |
Value
plotly object
Make predictions, linear model
Description
Make predictions, linear model
Usage
predict_lm(model, newdata, single)
Arguments
model |
a model object |
newdata |
data frame in which to look for variables with which to predict |
single |
logical: TRUE if single study experiment |
Value
data frame with predictions
Make predictions based on non-linear model, multiple studies
Description
Make predictions based on non-linear model, multiple studies
Usage
predict_nlm_multi(model, newdata, change_time)
Arguments
model |
an object of class brmsfit |
newdata |
data frame in which to look for variables with which to predict |
change_time |
data frame |
Value
data frame with predictions
Make predictions based on non-linear model, single study
Description
Make predictions based on non-linear model, single study
Usage
predict_nlm_single(model, newdata, change_time)
Arguments
model |
an object of class brmsfit |
newdata |
data frame in which to look for variables with which to predict |
change_time |
numeric |
Value
data frame with predictions
Make predictions
Description
Make predictions
Usage
predict_regr_model(model, df)
Arguments
model |
object of class brmsfit |
df |
data frame with classification results |
Value
data frame
Run the Shiny Application
Description
Run the Shiny Application
Usage
run_app(...)
Arguments
... |
additional options passed to shinyApp() |
Value
No return value, called for the shiny app interface
Fit nonlinear model - continuous hinge function
Description
Fit nonlinear model - continuous hinge function
Usage
run_nl_model(start, df_mod, formula, n_cores)
Arguments
start |
df with coefficients |
df_mod |
data of all variables used in the model |
formula |
an object of class brmsformula |
n_cores |
number of cores to use |
Value
object of class brmsfit
Set up a waiting screen
Description
Set up a waiting screen
Usage
set_waiter(header)
Arguments
header |
text to display on loading screen |
Value
object of a class waiter