how to make data available in Rshiny? - r

Question

I am trying to make data available in R shiny for survival analysis. In other words, some people might not have the access to upload data in R shiny. For those people, I want to provide some dataset so that they can use the data. Does anyone know how to make some data available in R shiny? This is the code:
library(shiny)
library(datasets)
library(survival)
options(shiny.maxRequestSize = 70 * 1024 ^ 2)
shinyUI(fluidPage(
titlePanel("Data Visualization"),
tabsetPanel(
tabPanel(
"Upload File",
titlePanel("Uploading Files"),
sidebarLayout(
sidebarPanel(
fileInput(
'file1',
'Choose RData File',
accept = c(".Rdata")
),
selectInput(
"dataset", "Select the dataset", choices = c("LUAD_RNASeq",
"LUSC_RNASeq")
)
),
mainPanel(
p("Head of the data"),
tableOutput('contents'))
)
),
tabPanel(
"Summary",
pageWithSidebar(
headerPanel('Summary of Variable'),
sidebarPanel(
# "Empty inputs" - they will be updated after the data is
uploaded
#selectInput('row', 'Row', ""),
selectInput('col', 'Column', "", selected = "")
),
mainPanel(
#p("Summary for the row"),
#verbatimTextOutput("row_summary"),
p("Summary for the column"),
verbatimTextOutput("col_summary")
)
)
),
tabPanel(
"Survival",
plotOutput("surPlot")
),
tabPanel(
"Scatter Plot",
pageWithSidebar(
headerPanel('Scatter Plot'),
sidebarPanel(
selectInput('col1', 'Column1', "", selected = ""),
selectInput('col2', 'Column2', "", selected = "")
),
mainPanel(
plotOutput("scaPlot")
)
)
),
tabPanel(
"Univariate Cox Regression",
pageWithSidebar(
headerPanel('Univariate Cox Regression'),
sidebarPanel(
selectInput('col3', 'Cox Variable', "", selected = "")
),
mainPanel(
verbatimTextOutput("reg_summary"),
plotOutput("reg_Plot")
)
)
),
tabPanel(
"AFT Regression",
pageWithSidebar(
headerPanel('Log normal AFT regression'),
sidebarPanel(
selectInput('col4', 'AFT Variable', "", selected = "")
),
mainPanel(
p("To do the AFT regression, the survival time can not be
zero"),
verbatimTextOutput("aft_summary")
#plotOutput("aft_Plot")
)
)
),
tabPanel(
"Glmnet Analysis",
p("To do Glmnetplot, the survival time can not be zero or NA
value"),
plotOutput("GlmnetPlot")
)
)
))
library(shiny)
library(datasets)
library(survival)
library(ggplot2)
library(survminer)
library(GGally)
library(glmnet)
library(rsconnect)
options(shiny.maxRequestSize = 70 * 1024 ^ 2)
load_obj <- function(f)
{
env <- new.env()
nm <- load(f, env)[1]
env[[nm]]
}
shinyServer(function(input, output, session) {
data <- reactive({
req(input$file1)
df<-load_obj(input$file1$datapath)$merged.dat
# Update inputs (you could create an observer with both updateSel...)
# You can also constraint your choices. If you wanted select only
numeric
# variables you could set "choices = sapply(df, is.numeric)"
# It depends on what do you want to do later on.
updateSelectInput(
session,
inputId = 'row',
label = 'Row',
choices = 1:nrow(df),
selected = 1
)
updateSelectInput(
session,
inputId = 'col',
label = 'Column',
choices = names(sapply(df, is.numeric)),
selected = names(df)[2]
)
updateSelectInput(
session,
inputId = 'col1',
label = 'Column1',
choices = names(sapply(df, is.numeric)),
selected = names(df)[4]
)
updateSelectInput(
session,
inputId = 'col2',
label = 'Column2',
choices = names(sapply(df, is.numeric)),
selected = names(df)[5]
)
updateSelectInput(
session,
inputId = 'col3',
label = 'Cox Variable',
choices = names(sapply(df, is.numeric)),
selected = names(df)[5]
)
updateSelectInput(
session,
inputId = 'col4',
label = 'AFT Variable',
choices = names(sapply(df, is.numeric)),
selected = names(df)[5]
)
return(df)
})
output$contents <- renderTable({
df<-data()
return (head(df[,1:10]))
})
output$row_summary <- renderPrint({
df<-data()
row<-as.numeric(df[input$row, 2:ncol(df)])
summary(row)
})
output$col_summary <- renderPrint({
df<-data()
column<-df[,input$col]
summary(column)
})
output$surPlot <- renderPlot({
plot.survival <- function(data)
{
ggsurv(survfit(
Surv(data$OS, data$status) ~ 1,
type = "kaplan-meier",
conf.type = "log-log"
),
main = "Survival Plot(K-M estimate)")
}
print(plot.survival(data()))
})
output$scaPlot <- renderPlot({
df<-data()
column1<-df[,input$col1]
column2<-df[,input$col2]
plot(column1, column2)
})
output$reg_summary <-renderPrint({
df=data()
column3=df[,input$col3]
res.cox=coxph(Surv(df$OS,df$status) ~ column3, data=df)
summary(res.cox)
})
output$reg_Plot <- renderPlot({
df=data()
column3=df[,input$col3]
res.cox=coxph(Surv(df$OS,df$status) ~ column3, data=df)
ggsurvplot(survfit(res.cox), palette = "#2E9FDF",ggtheme =
theme_minimal(),data = df)
})
output$aft_summary <-renderPrint({
df=data()
column4=df[,input$col4]
res.aft=survreg(Surv(df$OS,df$status) ~ column4, data=df,
dist="lognormal")
summary(res.aft)
})
output$GlmnetPlot <- renderPlot({
df=data()
y1=cbind(time=df$OS,status=df$statu)
x1 <- subset(df, select = -c(bcr,OS, status))
x1=data.matrix(x1, rownames.force = NA)
fit1=glmnet(x1,y1,family="cox")
plot(fit1)
})
# datasetInput <- eventReactive(input$update, {
# switch(input$dataset,
# "rock" = rock,
# "pressure" = pressure,
# "cars" = cars)
# })
# output$aft_Plot <- renderPlot({
# df=data()
# df=df[-c(which(df$OS==0)),]
# column4=df[,input$col4]
# res.aft=survreg(Surv(df$OS,df$status) ~ column4, data=df,
dist="lognormal")
# ggsurvplot(survfit(res.aft), color = "#2E9FDF",ggtheme =
theme_minimal(),data = df)
# })
})

Answer 1

If you have a CSV file with your data, include that in the folder your Shiny app is in, and call:
mydata <- read.csv("your_file.csv")
If you save your data as an R variable as an Rds:
mydata <- readRDS("your_file.Rds")