inst/Kronos_processing/app.R
In CL-CHEN-Lab/User_interface_for_Kronos_scRT: R-package and Shiny App interface for the analysis of single cell replication timing data
#options
options(stringsAsFactors = FALSE,
dplyr.summarise.inform=FALSE,
warn = 1,
scipen = 999)
#load operators
`%>%` = tidyr::`%>%`
`%dopar%` = foreach::`%dopar%`
`%do%` = foreach::`%do%`
#set theme for plots
ggplot2::theme_set(new = ggplot2::theme_bw())
#function to close app
jscode <- "shinyjs.closeWindow = function() { window.close(); }"
#number of available cores
maxCores = parallel::detectCores()
# Define UI
ui <- shinydashboard::dashboardPage(
title = 'Kronos scRT',
skin = 'yellow',
shinydashboard::dashboardHeader(title = shiny::span(
shiny::img(src = 'KronosLogo.png', width = '100%')
)),
shinydashboard::dashboardSidebar(
shinydashboard::sidebarMenu(
id = 'Sidebar',
shinydashboard::menuItem(text = "Home",
tabName = "Home"),
shinydashboard::menuItem(text = "Upload data",
tabName = "Upload"),
shinydashboard::menuItem(text = "Diagnostic",
tabName = 'Diagnostic'),
shinydashboard::menuItem(text = "Filter Cells",
tabName = "FilterCells"),
shinydashboard::menuItem(text = "Exit",
tabName = "Exit")
)
),
shinydashboard::dashboardBody(
#use shinyjs
shinyjs::useShinyjs(),
#to close window
shinyjs::extendShinyjs(text = jscode, functions = c("closeWindow")),
#use CSS
shiny::tags$head(
shiny::tags$link(rel = "stylesheet",
type = "text/css",
href = "custom.css")
),
#setting spinner
shinybusy::add_busy_spinner(
spin = "fading-circle",
position = 'bottom-right',
color = 'orange',
height = '200px',
width = '200px'
),
shinydashboard::tabItems(#home
{
shinydashboard::tabItem(
tabName = "Home",
shinydashboard::box(
width = 4,
height = '150',
title = 'Analysis name',
solidHeader = T,
background = 'black',
shiny::textInput(
inputId = 'Analysis_Name',
label = '',
value = 'Analysis'
)
),
shinydashboard::box(
width = 4,
height = '150',
title = 'Cores to use for the analysis',
solidHeader = T,
background = 'black',
align = 'center',
#cores
sliderInput(
width = '100%',
inputId = 'cores',
label = '',
value = trunc(maxCores / 2),
min = 1,
max = maxCores,
step = 1,
ticks = F
)
),
shinydashboard::box(
width = 4,
height = '150',
title = 'Output folder',
solidHeader = T,
background = 'black',
align = 'center',
shinyFiles::shinyDirButton(
id = 'Output_dir',
label = 'Output folder',
title = 'Output folder'
),
shiny::htmlOutput('Output_dir_out')
),
shinydashboard::box(
width = 4,
height = '150',
title = shiny::div(
'Chr prefix to use',
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'Depending on your reference genome, chromosomes or scaffolds could be named differently. Please, provide the prefix that precedes the chromosome or scaffold number. If no prefix is present this box has to be emptied.' , placement = 'right')
),
solidHeader = T,
background = 'black',
shiny::textInput(
inputId = 'chr_prefix',
label = '',
value = 'chr'
)
),
shiny::column(
width = 4,
height = '150',
align = 'center',
shiny::img(src = "Kronos_logo.png", height = 150)
),
shinydashboard::box(
width = 4,
height = '150',
title = shiny::div(
'Chr to include',
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'Chromosome to include in the analysis. You can define ranges using ":" and individual chromosomes using ",". Don\'t include the chr prefix!' , placement = 'right')
),
solidHeader = T,
background = 'black',
shiny::textInput(
inputId = 'chr_range',
label = '',
value = '1:22'
)
),
shinydashboard::box(
width = 4,
height = '150',
title = shiny::div(
'Chr size file',
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'For most genome assemblies this file can be downloaded from UCSC Genome Browser website.' , placement = 'right')
),
solidHeader = T,
background = 'black',
align = 'center',
shinyFiles::shinyFilesButton(
id = 'Chr_size',
label = 'Chr size file',
title = 'Chr size file',
multiple = F
),
shiny::htmlOutput('Chr_size_out', style = 'font-size: 16px;font-weight: bold;')
),
shinydashboard::box(
width = 4,
height = '150',
title = 'Final RT bin size',
solidHeader = T,
background = 'black',
align = 'center',
#cores
sliderInput(
width = '100%',
inputId = 'binsize',
label = '',
value = '0.5',
min = 0.1,
max = 2,
step = 0.05,
ticks = F,
post = 'Mb'
)
),
shinydashboard::box(
width = 4,
height = '150',
title = 'Apply settings',
solidHeader = T,
background = 'black',
align = 'center',
shinyjs::disabled(
shiny::actionButton(inputId = 'ApplySettings', label =
'Apply settings'))
)
)
},
#upload
{
shinydashboard::tabItem(
tabName = "Upload",
shinydashboard::box(
title = 'Upload samples',
solidHeader = T,
status = 'primary',
collapsible = T,
width = 12,
shiny::fluidRow(
shiny::column(
width = 3,
shinyFiles::shinyFilesButton(
id = 'PerCell_file',
label = shiny::div('PerCell file',
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = '_PerCell.csv file created by Kronos scRT processing.' , placement = 'right')
),
title = 'PerCell File',
multiple = F,
style = "width: 100%"
)
),
shiny::column(width = 9,
shiny::htmlOutput(outputId =
'PerCell_file_out'))
),
shiny::fluidRow(
shiny::column(
width = 3,
shinyFiles::shinyFilesButton(
id = 'scCN',
label = shiny::div('sc tracks file',
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = '_scCNV.tsv file created by Kronos scRT processing.' , placement = 'right')
),
title = 'sc tracks file',
multiple = F,
style = "width: 100%"
)
),
shiny::column(width = 9,
shiny::htmlOutput(outputId =
'scCN_out'))
),
shiny::fluidRow(
shiny::column(
width = 3,
shinyFiles::shinyFilesButton(
id = 'setting_file',
label = shiny::div(
'Setting file',
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'Optional. If this dataset has been used in a previous analysis the produced setting file can be reused.' , placement = 'right')
),
title = "Setting file",
multiple = F,
style = "width: 100%"
)
),
shiny::column(width = 9,
shiny::htmlOutput(outputId =
'setting_file_out'))
),
shiny::fluidRow(
shiny::column(
width = 3,
shinyFiles::shinyFilesButton(
id = 'whoiswho_file',
label = shiny::div(
"Who's who file",
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'Optional. A file containing experimental cell staging information. Such a file has to containing cell identifiers under the columns Cell and logical values under the columns S_Phase. (TRUE = a cell is in S phase, FALSE = a cell is in G1 or G2 phase)' , placement = 'right')
),
title = "Who's who file",
multiple = F,
style = "width: 100%"
)
),
shiny::column(width = 9,
shiny::htmlOutput(outputId = 'whoiswho_file_out'))
),
shiny::fluidRow(
shiny::column(
width = 6,
shiny::textInput(
inputId = ('FileName'),
value = 'Exp',
label = shiny::div(
"Basename",
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'This Name identifies each individual experiment' , placement = 'right')
),
width = '100%'
)
),
shiny::column(
width = 6,
shiny::textInput(
inputId = ('GroupName'),
value = 'Exp',
label = shiny::div(
"Group name",
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'If cells have been sequenced in different experiments, providing the same group name allows to merge them after normalization.' , placement = 'right')
),
width = '100%'
)
)
),
shiny::fluidRow(shiny::column(
width = 3,
shiny::actionButton(
inputId = 'Add_sample',
label = "Add sample",
width = '100%'
)
)),
shiny::fluidRow(shiny::column(
width = 12,
shiny::tableOutput('File_paths')
)),
shiny::fluidRow(shiny::column(
width = 3,
shiny::selectInput(
inputId = 'Remove_sample',
label = "Remove sample",
choices = 'Sample',
selected = 'Sample',
width = '100%'
)
))
),
shinydashboard::box(
title = shiny::div(
'Upload references',
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'Bulk RT references are optional. The program allows to upload maximum one reference per group.' , placement = 'right')
),
solidHeader = T,
status = 'primary',
collapsible = T,
collapsed = T,
width = 12,
shiny::fluidRow(
shiny::column(
width = 6,
shiny::selectInput(
inputId = 'reference_group',
label = 'group name',
choices = 'Select a group',
multiple = F,
selected = 'Select a group',
width = '100%'
)
),
shiny::column(
width = 6,
shiny::textInput(
inputId = 'reference_name',
value = 'Reference',
label = shiny::div(
'Reference name',
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'This name will be visualize on plots.' , placement = 'right')
),
width = '100%'
)
)
),
shiny::fluidRow(
shiny::column(
width = 3,
shinyFiles::shinyFilesButton(
id = 'reference_file',
label = "Reference RT file",
title = "Reference RT file",
multiple = F,
style = "width: 100%"
)
),
shiny::column(width = 9,
shiny::htmlOutput(outputId =
'reference_file_out'))
),
shiny::fluidRow(shiny::column(
width = 3,
shiny::actionButton(
inputId = 'Add_reference',
label = "Add reference",
width = '100%'
)
)),
shiny::fluidRow(shiny::column(
width = 12,
shiny::tableOutput('Reference_paths')
))
),
shiny::fluidRow(shiny::column(
width = 3,
shinyjs::disabled(
shiny::actionButton(
inputId = 'Done_upload',
label = 'Done',
width = '100%'
)
)
))
)
},
#Diagnostic
{
shinydashboard::tabItem(tabName = 'Diagnostic',
shiny::uiOutput('Diagnostic_UI'),
shiny::fluidRow(shiny::column(
width = 3,
shiny::actionButton(
inputId = 'Diagnostic_done',
label = "Start RT analysis",
width = '100%'
)
)))
},
#Filter Cells
{
shinydashboard::tabItem(
tabName = 'FilterCells',
shiny::fluidRow(shiny::column(
width = 6,
offset = 3,
shiny::sliderInput(
width = '100%',
inputId = 'min_correlation_matrix',
label = 'Min correlation to be kept',
min = 0,
max = 1,
step = 0.01,
value = 0.25
)
)),
shiny::fluidRow(
shiny::column(width = 6,
plotly::plotlyOutput('plot1_RT',width = '100%',height = 'auto')),
shiny::column(width = 6,
plotly::plotlyOutput('plot2_RT',width = '100%',height = 'auto'))
),
shiny::fluidRow(shiny::column(
width = 3,
shiny::actionButton(
inputId = 'RT_next',
label = "Finalize",
width = '100%'
)
)),
shiny::fluidRow(
shiny::column(
width = 12,
align = 'center',
shiny::htmlOutput('END')
)
)
)
})
)
)
server <- function(input, output, session) {
#variables
variables = shiny::reactiveValues(
setting_file = '',
File_paths = dplyr::tibble(),
Reference_paths = dplyr::tibble(),
PerCell_file = '',
whoiswho_file = '',
PerCell_file_error = '',
first_line_Percell = NULL,
scCN = '',
scCN_error = '',
first_line_CNV = NULL,
reference_file = '',
groups = '',
groups_with_a_reference = '',
Chr_size = '<font color=\"#FF0000\"><b> This is a mandatory input.</b></font>',
roots = c(
shinyFiles::getVolumes()(),
Home = Sys.getenv("HOME"),
OutputFolder = file.path(Sys.getenv("HOME"))
),
to_keep = NULL,
Save__scPlots = F,
Save__BinRep = F
)
#store diagnostic module info
diagnostic_module_ls = shiny::reactiveValues(ui = list(),
server = list())
#store data
data = shiny::reactiveValues(
PerCell = dplyr::tibble(),
scCN = dplyr::tibble(),
whoswho = dplyr::tibble(),
setting = dplyr::tibble(),
reference = dplyr::tibble(),
variability = dplyr::tibble(),
scRT = dplyr::tibble(),
G1G2 = dplyr::tibble()
)
#load
shiny::observe({
#add tab items in diagnostic
output$Diagnostic_UI = shiny::renderUI({
diagnostic_module_ls$ui
})
})
#stop app when the session ends
session$onSessionEnded(function() {
shiny::stopApp()
})
#home
{
shiny::observe({
shinyFiles::shinyFileChoose(
input = input,
id = 'Chr_size',
session = session,
roots = variables$roots,
defaultRoot = 'Home'
)
shinyFiles::shinyDirChoose(
input = input,
id = 'Output_dir',
session = session,
roots = variables$roots,
defaultRoot = 'Home'
)
shiny::observeEvent(input$Analysis_Name, {
updateTextInput(
session = session,
inputId = 'Analysis_Name',
value = stringr::str_replace_all(
string = input$Analysis_Name,
pattern = ' ',
replacement = '_'
)
)
})
output$Output_dir_out <-
renderText(paste(
'<H4><b>',
file.path(variables$roots['OutputFolder'], input$Analysis_Name),
'</H4></b>'
))
if (!stringr::str_detect(string = variables$Chr_size,
pattern = '<font color=\"#FF0000\">')) {
shinyjs::enable('Upload_chr_size')
} else{
shinyjs::disable('Upload_chr_size')
}
})
#load chrs_size
shiny::observeEvent(input$Chr_size, {
if (!is.numeric(input$Chr_size)) {
#get path
variables$Chr_size = dplyr::tibble(
shinyFiles::parseFilePaths(
roots = variables$roots ,
selection = input$Chr_size
)
) %>% dplyr::pull(datapath)
variables$Chr_size = Kronos.scRT::right_format(
file_path = variables$Chr_size,
delim = '\t',
columns_to_check = 2,
wrong_message = paste0(
'<b>',
variables$Chr_size,
'<font color=\"#FF0000\"> does not have the right format. </b></font>'
),
rigth_message = variables$Chr_size
)
} else{
variables$Chr_size = '<font color=\"#FF0000\"><b> This is a mandatory input.</b></font>'
}
})
shiny::observe(output$Chr_size_out <-
renderText(variables$Chr_size))
shiny::observeEvent(input$Output_dir, {
if (!is.numeric(input$Output_dir)) {
variables$roots = c(
shinyFiles::getVolumes()(),
Home = Sys.getenv("HOME"),
OutputFolder = parseDirPath(
roots = variables$roots,
selection = input$Output_dir
)
)
}
})
shiny::observeEvent(input$Chr_size,{
if (!is.numeric(input$Chr_size) & input$ApplySettings == 0) {
shinyjs::enable('ApplySettings')
} else{
shinyjs::disable('ApplySettings')
}
})
shiny::observeEvent(input$ApplySettings, {
if (input$ApplySettings == 1) {
shinyjs::disable('ApplySettings')
shinyjs::disable('binsize')
shinyjs::disable('Chr_size')
shinyjs::disable('chr_range')
shinyjs::disable('chr_prefix')
shinyjs::disable('Output_dir')
shinyjs::disable('cores')
shinyjs::disable('Analysis_Name')
#move to next tab
shinydashboard::updateTabItems(session = session,
inputId = 'Sidebar',
selected = 'Upload')
}
})
}
#load data
{
#samples
{
shiny::observe({
shinyFiles::shinyFileChoose(
input = input,
id = 'PerCell_file',
roots = variables$roots,
session = session,
defaultRoot = 'Home'
)
shinyFiles::shinyFileChoose(
input = input,
id = 'scCN',
roots = variables$roots,
session = session,
defaultRoot = 'Home'
)
shinyFiles::shinyFileChoose(
input = input,
id = 'whoiswho_file',
roots = variables$roots,
session = session,
defaultRoot = 'Home'
)
shinyFiles::shinyFileChoose(
input = input,
id = 'setting_file',
roots = variables$roots,
session = session,
defaultRoot = 'Home'
)
})
#activate deactivate Add input button
shiny::observe({
if (variables$PerCell_file == '' | variables$scCN == '' |
stringr::str_detect(string = variables$PerCell_file, pattern = ' does not have the right format') |
stringr::str_detect(string = variables$scCN, pattern = ' does not have the right format') |
stringr::str_detect(string = variables$whoiswho_file, pattern = ' does not have the right format') |
input$ApplySettings != 1) {
shinyjs::disable(id = 'Add_sample')
output$PerCell_file_out = shiny::renderText(variables$PerCell_file)
output$scCN_out = shiny::renderText(variables$scCN)
} else{
if (all(!c('basename', 'group') %in% names(variables$first_line_Percell)) &
all(!c('basename', 'group') %in% names(variables$first_line_CNV))) {
#rest outputs if they had changed
output$PerCell_file_out = shiny::renderText(variables$PerCell_file)
output$scCN_out = shiny::renderText(variables$scCN)
shinyjs::enable(id = 'Add_sample')
} else if (all(c('basename', 'group') %in% names(variables$first_line_Percell)) &
all(c('basename', 'group') %in% names(variables$first_line_CNV))) {
if (variables$first_line_Percell$basename == variables$first_line_CNV$basename &
variables$first_line_Percell$group == variables$first_line_CNV$group) {
#update file name and group name if they match
updateTextInput(
session = session,
inputId = 'FileName',
value = variables$first_line_Percell$basename
)
updateTextInput(
session = session,
inputId = 'GroupName',
value = variables$first_line_Percell$group
)
#rest outputs if they had changed
output$PerCell_file_out = shiny::renderText(variables$PerCell_file)
output$scCN_out = shiny::renderText(variables$scCN)
shinyjs::enable(id = 'Add_sample')
} else{
#set warning
output$PerCell_file_out = shiny::renderText(
'<font color=\"#FF0000\"> Basename and Group do not match between PerCell and sc tracks files!!! </b></font>'
)
output$scCN_out = shiny::renderText(
'<font color=\"#FF0000\"> Basename and Group do not match between PerCell and sc tracks files!!! </b></font>'
)
shinyjs::disable(id = 'Add_sample')
}
} else{
output$PerCell_file_out = shiny::renderText(
'<font color=\"#FF0000\"> Basename and Group do not match between PerCell and sc tracks files!!! </b></font>'
)
output$scCN_out = shiny::renderText(
'<font color=\"#FF0000\"> Basename and Group do not match between PerCell and sc tracks files!!! </b></font>'
)
shinyjs::disable(id = 'Add_sample')
}
}
#activate button to srat RT analysis
if (nrow(variables$File_paths) > 0) {
shinyjs::enable(id = 'Diagnostic_done')
shinyjs::enable(id = 'Done_upload')
} else{
shinyjs::disable(id = 'Diagnostic_done')
shinyjs::disable(id = 'Done_upload')
}
})
#move to next page
shiny::observeEvent(input$Done_upload, {
shinydashboard::updateTabItems(session = session,
inputId = 'Sidebar',
selected = 'Diagnostic')
})
#replace ' ' with '_'
shiny::observeEvent(input$FileName, {
updateTextInput(
session = session,
inputId = 'FileName',
value = stringr::str_replace_all(
string = input$FileName,
pattern = ' ',
replacement = '_'
)
)
})
shiny::observeEvent(input$GroupName, {
updateTextInput(
session = session,
inputId = 'GroupName',
value = stringr::str_replace_all(
string = input$GroupName,
pattern = ' ',
replacement = '_'
)
)
})
# get PerCell_file path and check if it has the right format
shiny::observeEvent(input$PerCell_file, {
if (!is.numeric(input$PerCell_file)) {
#get path
variables$PerCell_file = dplyr::tibble(
shinyFiles::parseFilePaths(
roots = variables$roots ,
selection = input$PerCell_file
)
) %>% dplyr::pull(datapath)
#check if it is the right file
variables$PerCell_file = Kronos.scRT::right_format(
file_path = variables$PerCell_file,
delim = ',',
columns_to_check = c(
"Cell",
"normalized_dimapd",
"mean_ploidy",
"ploidy_confidence",
"is_high_dimapd",
"is_noisy",
"coverage_per_1Mbp"
),
wrong_message = paste0(
'<b>',
variables$PerCell_file,
'<font color=\"#FF0000\"> does not have the right format. </b></font>'
),
rigth_message = variables$PerCell_file
)
#read line one and look for group and basename
if (stringr::str_detect(
string = variables$PerCell_file,
pattern = ' does not have the right format',
negate = T
)) {
variables$first_line_Percell = readr::read_csv(variables$PerCell_file, n_max = 1)
}
} else{
variables$PerCell_file = ''
}
})
# get scCN path and check if it has the right format
shiny::observeEvent(input$scCN, {
if (!is.numeric(input$scCN)) {
#get path
variables$scCN = dplyr::tibble(
shinyFiles::parseFilePaths(roots = variables$roots ,
selection = input$scCN)
) %>% dplyr::pull(datapath)
#check if it is the right file
variables$scCN = Kronos.scRT::right_format(
file_path = variables$scCN,
delim = '\t',
columns_to_check = c('chr',
'start',
'end' ,
'copy_number',
'reads' ,
'Cell'),
wrong_message = paste0(
'<b>',
variables$scCN,
'<font color=\"#FF0000\"> does not have the right format. </b></font>'
),
rigth_message = variables$scCN
)
#read line one and look for group and basename
if (stringr::str_detect(
string = variables$scCN,
pattern = ' does not have the right format',
negate = T
)) {
variables$first_line_CNV = readr::read_tsv(variables$scCN, n_max = 1)
}
} else{
variables$scCN = ''
}
})
# get whoiswho path and check if it has the right format
shiny::observeEvent(input$whoiswho_file, {
if (!is.numeric(input$whoiswho_file)) {
#get path
variables$whoiswho_file = dplyr::tibble(
shinyFiles::parseFilePaths(
roots = variables$roots ,
selection = input$whoiswho_file
)
) %>% dplyr::pull(datapath)
#check if it is the right file
variables$whoiswho_file = Kronos.scRT::right_format(
file_path = variables$whoiswho_file,
delim = '\t',
columns_to_check = c('Cell', 'Phase'),
wrong_message = paste0(
'<b>',
variables$whoiswho_file,
'<font color=\"#FF0000\"> does not have the right format. </b></font>'
),
rigth_message = variables$whoiswho_file
)
output$whoiswho_file_out = shiny::renderText(variables$whoiswho_file)
} else{
variables$whoiswho_file = ''
output$whoiswho_file_out = shiny::renderText(variables$whoiswho_file)
}
})
# get setting path and check if it has the right format
shiny::observeEvent(input$setting_file, {
if (!is.numeric(input$setting_file)) {
#get path
variables$setting_file = dplyr::tibble(
shinyFiles::parseFilePaths(
roots = variables$roots ,
selection = input$setting_file
)
) %>% dplyr::pull(datapath)
#check if it is the right file
variables$setting_file = Kronos.scRT::right_format(
file_path = variables$setting_file,
delim = '\t',
columns_to_check = c(
'threshold_Sphase' ,
'threshold_G1G2phase' ,
'Sphase_first_part',
'Sphase_second_part' ,
'RPMPH_TH',
'RPM_TH'
),
wrong_message = paste0(
'<b>',
variables$setting_file,
'<font color=\"#FF0000\"> does not have the right format. </b></font>'
),
rigth_message = variables$setting_file
)
output$setting_file_out = shiny::renderText(variables$setting_file)
} else{
variables$setting_file = ''
output$setting_file_out = shiny::renderText(variables$setting_file)
}
})
#add samples to dataframe
shiny::observeEvent(input$Add_sample, {
shinyjs::disable(id = 'Add_sample')
shinyjs::disable(id = 'Remove_sample')
#create directory if it does not exist
if (!dir.exists(file.path(variables$roots['OutputFolder'], input$Analysis_Name))) {
dir.create(file.path(variables$roots['OutputFolder'], input$Analysis_Name),
recursive = T)
}
#check if it is a new group
if (!(
input$GroupName %in% variables$groups |
input$GroupName %in% variables$groups_with_a_reference
)) {
variables$groups = c(variables$groups, input$GroupName)
}
#new line
newline = dplyr::tibble(
PerCellFile = basename(variables$PerCell_file),
scCNV = basename(variables$scCN),
basename = input$FileName,
group = input$GroupName,
WhoSWho = ifelse(
!is.numeric(input$whoiswho_file),
basename(variables$whoiswho_file),
''
),
Setting = ifelse(
!is.numeric(input$setting_file),
basename(variables$setting_file),
''
)
)
#load data
#if whoiswho_file's been provided
if (!is.numeric(input$whoiswho_file)) {
data$PerCell = rbind(
data$PerCell,
readr::read_csv(variables$PerCell_file, col_types = 'cnnnlln') %>%
dplyr::mutate(
basename = input$FileName,
group = input$GroupName
)
) %>%
dplyr::inner_join(readr::read_tsv(variables$whoiswho_file, col_types = 'cc'),
by = c("Cell")) %>%
dplyr::mutate(
is_high_dimapd = ifelse(Phase == 'S', T, F),
is_noisy = ifelse(is_high_dimapd, T, is_noisy)
) %>%
dplyr::select(-Phase)
} else{
data$PerCell = rbind(
data$PerCell,
readr::read_csv(variables$PerCell_file, col_types = 'cnnnlln') %>%
dplyr::mutate(
basename = input$FileName,
group = input$GroupName
)
)
}
data$scCN = rbind(
data$scCN,
readr::read_tsv(variables$scCN, col_types = 'ccnnnn') %>%
dplyr::mutate(
basename = input$FileName,
group = input$GroupName
)
)
if (!is.numeric(input$whoiswho_file) &
variables$whoiswho_file != '') {
data$whoswho = rbind(
data$whoswho,
readr::read_tsv(variables$whoiswho_file, col_types = 'cc') %>%
dplyr::mutate(
basename = input$FileName,
group = input$GroupName
)
)
}
#if setting_file's been provided
if (!is.numeric(input$setting_file) &
variables$setting_file != '') {
data$setting = rbind(
data$setting,
readr::read_tsv(variables$setting_file, col_types = 'nnnnnnncc') %>%
dplyr::mutate(
basename = input$FileName,
group = input$GroupName
)
)
}
#add line to dataframe
variables$File_paths = rbind(variables$File_paths,
newline)
#call diagnostic module
diagnostic_module_ls$ui[[paste0(input$FileName)]] = diagnostic_ui(id = input$FileName)
diagnostic_module_ls$server[[paste0(input$FileName)]] = diagnostic_server(
id = input$FileName,
PerCell = data$PerCell %>%
dplyr::filter(basename ==
input$FileName),
#if setting is not there return empty tibble
Setting = tryCatch(
data$setting %>%
dplyr::filter(basename ==
input$FileName),
error =
function(x)
return(dplyr::tibble())
),
Cores = input$cores,
OutFolder = file.path(variables$roots['OutputFolder'], input$Analysis_Name)
)
#update Remove_sample
shiny::updateSelectInput(
inputId = 'Remove_sample',
choices = c('Sample', variables$File_paths$basename)
)
# reset inputs
shiny::updateTextInput(
session = session,
inputId = 'FileName',
value = paste('Exp', input$Add_sample, sep = '_')
)
shiny::updateTextInput(
session = session,
inputId = 'GroupName',
value = paste('Exp', input$Add_sample, sep = '_')
)
variables$PerCell_file = ''
variables$scCN = ''
variables$whoiswho_file = ''
variables$setting_file = ''
output$scCN_out = shiny::renderText('')
output$PerCell_file_out = shiny::renderText('')
output$whoiswho_file_out = shiny::renderText('')
output$setting_file_out = shiny::renderText('')
output$File_paths = shiny::renderTable(variables$File_paths)
#update groups for references
shiny::updateSelectInput(
inputId = 'reference_group',
choices = c('Select a group',
variables$groups),
selected = 'Select a group'
)
#reactivate remove sample
shinyjs::enable('Remove_sample')
})
#remove samples
shiny::observeEvent(input$Remove_sample, {
if (input$Remove_sample != 'Sample') {
#remove
variables$File_paths = variables$File_paths %>%
dplyr::filter(basename != input$Remove_sample)
# stop module
diagnostic_module_ls$ui[[input$Remove_sample]] = NULL
diagnostic_module_ls$server[[input$Remove_sample]] = NULL
#update Remove_sample
shiny::updateSelectInput(
inputId = 'Remove_sample',
choices = c('Sample', variables$File_paths$basename),
selected = 'Sample'
)
#update groups for Reference
variables$groups = variables$groups[variables$groups %in% unique(variables$File_paths$group)]
shiny::updateSelectInput(
inputId = 'reference_group',
choices = c('Select a group',
variables$groups),
selected = 'Select a group'
)
}
})
shiny::observeEvent(input$Diagnostic_done, {
#disable button
shinyjs::disable('Diagnostic_done')
#find all the setting files created by diagnostic
files_to_load = list.files(
path = file.path(variables$roots['OutputFolder'], input$Analysis_Name),
pattern = '_settings.txt',
full.names = T
)
# select those with the right format
files_to_load = files_to_load[unlist(
lapply(
files_to_load,
Kronos.scRT::right_format,
delim = '\t',
columns_to_check = c(
'threshold_Sphase' ,
'threshold_G1G2phase' ,
'Sphase_first_part',
'Sphase_second_part' ,
'Ploidy',
'RPMPH_TH',
'RPM_TH',
'basename',
'group'
),
wrong_message = F,
rigth_message = T
)
)]
#upload new setting file
data$setting = Kronos.scRT::load_multiple_df(dirs = files_to_load,
delim = '\t',
col_types = 'nnnnnnncc')
########RT Start first part
{
variables$chr_list = paste0(input$chr_prefix,
unlist(
Kronos.scRT::String_to_Range(x = stringr::str_split(input$chr_range, ',')[[1]])
))
#upload chr sizes
data$Chr_size = readr::read_tsv(
variables$Chr_size,
col_names = c('chr', 'size'),
col_types = 'cn'
) %>%
dplyr::filter(chr %in% variables$chr_list) %>%
dplyr::mutate(chr = factor(x = chr, levels = variables$chr_list)) %>%
tidyr::drop_na()
variables$chr_list = variables$chr_list[variables$chr_list %in% unique(data$Chr_size$chr)]
#Apply Settings to PerCell
data$PerCell = Kronos.scRT::AdjustPerCell(
PerCell = data$PerCell ,
Settings = data$setting,
Basename_leves = variables$File_paths$basename ,
Group_leves = unique(variables$File_paths$group)
)
#Correct scCN based on PerCell
data$scCN = Kronos.scRT::AdjustCN(
scCN = data$scCN,
PerCell = data$PerCell ,
Basename_leves = variables$File_paths$basename ,
Group_leves = unique(variables$File_paths$group),
Chr_filter = variables$chr_list
)
#create bins
bins = Kronos.scRT::GenomeBinning(
Chr_size = data$Chr_size,
size = input$binsize * 10 ^ 6,
Chr_filter = variables$chr_list,
Cores = input$cores
)
#rebin data
data$signal_smoothed = Kronos.scRT::Rebin(
PerCell = data$PerCell,
scCN = data$scCN,
Bins = bins,
Sphase = T
)
G1G2_smoothed = Kronos.scRT::Rebin(
PerCell = data$PerCell,
scCN = data$scCN,
Bins = bins,
Sphase = F
)
#calculate background
median_G1G2_profile = Kronos.scRT::BackGround(G1_scCN = G1G2_smoothed)
# create single G1/G2 single cell file
G1G2_smoothed = Kronos.scRT::Replication_state(
Samples = G1G2_smoothed,
background = median_G1G2_profile,
Chr_filter = variables$chr_list,
cores = input$cores
)
#reshape and save
G1G2_smoothed = G1G2_smoothed %>%
dplyr::select(
chr,
start,
end,
CN,
background,
CN_bg,
th,
Rep,
PercentageReplication,
Cell,
basename,
group,
newIndex
)
G1G2_smoothed %>%
readr::write_tsv(
file = file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
paste0(
input$Analysis_Name,
'_G1_G2_single_cells_CNV_',
input$binsize,
'Mb.tsv'
)
),
col_names = T
)
data$G1G2 = G1G2_smoothed
if (nrow(data$reference) != 0) {
# rebin reference RT
data$reference = Kronos.scRT::RebinRT(
RT = data$reference,
Bins = bins,
Chr_filter = variables$chr_list
)
#write output
readr::write_tsv(
x = data$reference,
file = file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
paste0(
input$Analysis_Name,
'_reference_replication_timing_',
input$binsize,
'Mb.tsv'
)
),
col_names = T
)
}
#Calculate replication state S
data$signal_smoothed = Kronos.scRT::Replication_state(
Samples = data$signal_smoothed,
background = median_G1G2_profile,
Chr_filter = variables$chr_list,
cores = input$cores
)
# remove control track
rm('median_G1G2_profile')
#matrix for the correlation
data$signal_smoothed = data$signal_smoothed %>%
dplyr::ungroup() %>%
dplyr::arrange(group, newIndex) %>%
tidyr::unite(index, c(group, newIndex), sep = ' _ ') %>%
dplyr::mutate(index = factor(index, levels = unique(index)))
mat = data$signal_smoothed %>%
tidyr::unite(pos, c(chr, start), sep = ':') %>%
dplyr::mutate(Rep = as.numeric(Rep)) %>%
dplyr::select(pos, index, Rep) %>%
tidyr::spread(key = index, value = Rep) %>%
tibble::column_to_rownames('pos') %>%
dplyr::filter(complete.cases(.)) %>%
as.matrix()
#index
variables$Index = colnames(mat)
#correlation similarity distance
variables$results = 1 - as.matrix(ade4::dist.binary(
t(mat),
method = 2,
diag = T,
upper = T
))
variables$basenames = dplyr::tibble(Group = stringr::str_remove(colnames(mat), ' _ [0-9]{1,10}$'))
#write matrix and plot heatmap before filtering
saveRDS(
object = variables$results,
file = file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
paste0(
input$Analysis_Name,
'_correlation_per_cell_before_filtering.rds'
)
)
)
#prepare color patterns
variables$color = grDevices::colorRampPalette(
colors = c(
"#00204DFF",
"#233E6CFF",
"#575C6DFF",
"#7C7B78FF",
"#A69D75FF",
"#D3C164FF",
"#FFEA46FF"
)
)
output$plot1_RT <- plotly::renderPlotly(
{heatmaply::heatmaply(
x = variables$results,
colors = variables$color,
dendrogram = F,
showticklabels = F,
row_side_colors = variables$basenames,
col_side_colors = variables$basenames,
limits = c(0, 1)
) %>%
plotly::layout(
showlegend = FALSE,
legend = FALSE,
annotations = list(visible = FALSE)
)}
)
variables$to_keep = foreach::foreach(i = unique(variables$basenames$Group)) %do% {
sub_mat = variables$results[variables$basenames$Group == i, variables$basenames$Group == i]
diag(sub_mat) = 0
! matrixStats::rowQuantiles(x = sub_mat,
probs = 0.60,
na.rm = T) <= input$min_correlation_matrix
}
variables$to_keep = unlist(variables$to_keep)
variables$results_after_filtering = variables$results[variables$to_keep, variables$to_keep]
variables$basenames_after_filtering = variables$basenames[variables$to_keep,]
output$plot2_RT <- plotly::renderPlotly({
heatmaply::heatmaply(
x = variables$results_after_filtering,
colors = variables$color,
dendrogram = F,
showticklabels = F,
row_side_colors = variables$basenames_after_filtering,
col_side_colors = variables$basenames_after_filtering,
limits = c(0, 1)
) %>%
plotly::layout(
showlegend = FALSE,
legend = FALSE,
annotations = list(visible = FALSE)
)}
)
#bring the user to FilterCells
shinydashboard::updateTabItems(session = session,
inputId = 'Sidebar',
selected = 'FilterCells')
}
########RT end first part
})
###filter
shiny::observeEvent(input$min_correlation_matrix, {
if (!is.null(variables$to_keep)) {
variables$to_keep = foreach::foreach(i = unique(variables$basenames$Group)) %do% {
sub_mat = variables$results[variables$basenames$Group == i, variables$basenames$Group == i]
diag(sub_mat) = 0
! matrixStats::rowQuantiles(x = sub_mat,
probs = 0.60,
na.rm = T) <= input$min_correlation_matrix
}
variables$to_keep = unlist(variables$to_keep)
variables$results_after_filtering = variables$results[variables$to_keep, variables$to_keep]
variables$basenames_after_filtering = variables$basenames[variables$to_keep,]
output$plot2_RT <- plotly::renderPlotly(
heatmaply::heatmaply(
x = variables$results_after_filtering,
colors = variables$color,
dendrogram = F,
showticklabels = F,
row_side_colors = variables$basenames_after_filtering,
col_side_colors = variables$basenames_after_filtering,
limits = c(0, 1)
) %>%
plotly::layout(
showlegend = FALSE,
legend = FALSE,
annotations = list(visible = FALSE)
)
)
}
})
########RT second part start
shiny::observeEvent(input$RT_next, {
if (input$RT_next == 1) {
shinyjs::disable('RT_next')
#write matrix and plot heatmap after filtering
saveRDS(
object = variables$results_after_filtering,
file = file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
paste0(
input$Analysis_Name,
'_correlation_per_cell_after_filtering.rds'
)
)
)
variables$Index = variables$Index[variables$to_keep]
#filter out samples that don't correlate and save
data$signal_smoothed = data$signal_smoothed %>%
dplyr::filter(index %in% variables$Index) %>%
tidyr::separate(index, c('group', 'index'), sep = ' _ ') %>%
dplyr::mutate(group = factor(group, level = unique(variables$File_paths$group)))
rep_percentage = data$signal_smoothed %>%
dplyr::group_by(Cell, basename, group, index) %>%
dplyr::summarise(Rep_percentage = mean(Rep))
#new index
new_index_list = rep_percentage %>%
dplyr::ungroup() %>%
dplyr::arrange(Rep_percentage) %>%
dplyr::group_by(group) %>%
dplyr::mutate(newIndex = 1:dplyr::n()) %>%
dplyr::arrange(group, newIndex) %>%
dplyr::select(oldIndex = index, newIndex, Cell, basename, group)
data$signal_smoothed = data$signal_smoothed %>%
dplyr::ungroup() %>%
dplyr::inner_join(new_index_list,
by = c('Cell', 'index' = 'oldIndex', 'basename', 'group')) %>%
dplyr::select(
chr,
start,
end,
CN,
background,
CN_bg,
th,
Rep,
PercentageReplication,
Cell,
basename,
group,
newIndex
)
readr::write_tsv(
x = data$signal_smoothed,
file = file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
paste0(
input$Analysis_Name,
'_single_cells_CNV_',
input$binsize,
'Mb.tsv'
)
),
col_names = T
)
#select used data and save the new per cell files
used_cells = rbind(
new_index_list %>%
dplyr::select(Cell, basename, group) %>%
dplyr::ungroup(),
data$PerCell %>%
dplyr::filter(Type == 'G1/G2-phase cells') %>%
dplyr::select(Cell, basename, group) %>%
dplyr::ungroup()
)
data$PerCell = data$PerCell %>%
dplyr::inner_join(used_cells,
Joining,
by = c("Cell", "basename", "group"))
#create folder
if (!dir.exists(
file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
'Cells_used_in_the_analysis_info'
)
)) {
dir.create(
file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
'Cells_used_in_the_analysis_info'
),
recursive = T
)
}
#isolate for foreach
PerCell = shiny::isolate(data$PerCell)
bs = foreach::foreach(i = unique(PerCell$basename)) %do% {
PerCell %>%
dplyr::filter(basename == i) %>%
dplyr::select(
Cell,
normalized_dimapd,
mean_ploidy,
ploidy_confidence,
is_high_dimapd,
is_noisy,
coverage_per_1Mbp
) %>%
readr::write_csv(
file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
'Cells_used_in_the_analysis_info',
paste0(
input$Analysis_Name,
i,
'_per_Cell_summary_metrics.csv'
)
)
)
i
}
rm('bs')
rm('PerCell')
rm('new_index_list')
#calculate psudo bulk
data$scRT = Kronos.scRT::pseudoBulkRT(data$signal_smoothed)
readr::write_tsv(
x = data$scRT,
file = file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
paste0(
input$Analysis_Name,
'_calculated_replication_timing_',
input$binsize,
'Mb.tsv'
)
),
col_names = T
)
#merge reference and scRT
data$RTs = rbind(data$scRT %>% dplyr::ungroup(),
data$reference %>% dplyr::ungroup())
#prepare for twidth
data$variability = Kronos.scRT::Variability(S_scCN = data$signal_smoothed,
scRT = data$scRT)
readr::write_tsv(
x = data$variability,
file = file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
paste0(input$Analysis_Name,
'_scRT_variability.tsv')
),
col_names = T
)
output$END <- shiny::renderText({'<h3><font color=\"#FF0000\"> DONE! </b></font></h3>' })
}
})
########RT second part end
}
#reference
{
shiny::observe({
shinyFiles::shinyFileChoose(
input = input,
id = 'reference_file',
roots = variables$roots,
session = session,
defaultRoot = 'Home'
)
if (file.exists(variables$reference_file) &
input$reference_group != 'Select a group') {
shinyjs::enable('Add_reference')
} else{
shinyjs::disable('Add_reference')
}
})
#select and check reference file
shiny::observeEvent(input$reference_file, {
if (!is.numeric(input$reference_file)) {
#get path
variables$reference_file = dplyr::tibble(
shinyFiles::parseFilePaths(
roots = variables$roots ,
selection = input$reference_file
)
) %>% dplyr::pull(datapath)
#check if it is the right file
variables$reference_file = Kronos.scRT::right_format(
file_path = variables$reference_file,
delim = '\t',
columns_to_check = 4,
wrong_message = paste0(
'<b>',
variables$reference_file,
'<font color=\"#FF0000\"> does not have the right format. </b></font>'
),
rigth_message = variables$reference_file
)
output$reference_file_out = shiny::renderText(variables$reference_file)
} else{
variables$reference_file = ''
output$reference_file_out = shiny::renderText(variables$reference_file)
}
})
#replace ' ' with '_' for reference name
shiny::observeEvent(input$reference_name, {
updateTextInput(
session = session,
inputId = 'reference_name',
value = stringr::str_replace_all(
string = input$reference_name,
pattern = ' ',
replacement = '_'
)
)
})
#add reference
shiny::observeEvent(input$Add_reference, {
shinyjs::disable('Add_reference')
data$reference = rbind(
data$reference,
readr::read_tsv(
variables$reference_file,
col_names = c('chr', 'start', 'end', 'RT'),
col_types = 'cnnn'
) %>%
dplyr::mutate(
basename = input$reference_name,
group = input$reference_group
)
)
variables$Reference_paths = rbind(
variables$Reference_paths,
dplyr::tibble(
File = basename(variables$reference_file),
basename = input$reference_name,
group = input$reference_group
)
)
output$Reference_paths = shiny::renderTable(variables$Reference_paths)
variables$reference_file = ''
output$reference_file_out = shiny::renderText('')
variables$groups_with_a_reference = c(variables$groups_with_a_reference,
input$reference_group)
variables$groups = variables$groups[!variables$groups %in% variables$groups_with_a_reference]
#update groups for Reference
shiny::updateSelectInput(
inputId = 'reference_group',
choices = c('Select a group',
variables$groups),
selected = 'Select a group'
)
updateTextInput(inputId = 'reference_name',
value = paste0('Reference_', input$Add_reference))
})
}
}
####exit
{
shiny::observe({
if (input$Sidebar == 'Exit') {
shinyjs::js$closeWindow()
shiny::stopApp()
}
})
}
}
# Run the application
shiny::shinyApp(ui = ui,
server = server)
CL-CHEN-Lab/User_interface_for_Kronos_scRT documentation built on Aug. 1, 2022, 2:08 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
#options
options(stringsAsFactors = FALSE,
dplyr.summarise.inform=FALSE,
warn = 1,
scipen = 999)
#load operators
`%>%` = tidyr::`%>%`
`%dopar%` = foreach::`%dopar%`
`%do%` = foreach::`%do%`
#set theme for plots
ggplot2::theme_set(new = ggplot2::theme_bw())
#function to close app
jscode <- "shinyjs.closeWindow = function() { window.close(); }"
#number of available cores
maxCores = parallel::detectCores()
# Define UI
ui <- shinydashboard::dashboardPage(
title = 'Kronos scRT',
skin = 'yellow',
shinydashboard::dashboardHeader(title = shiny::span(
shiny::img(src = 'KronosLogo.png', width = '100%')
)),
shinydashboard::dashboardSidebar(
shinydashboard::sidebarMenu(
id = 'Sidebar',
shinydashboard::menuItem(text = "Home",
tabName = "Home"),
shinydashboard::menuItem(text = "Upload data",
tabName = "Upload"),
shinydashboard::menuItem(text = "Diagnostic",
tabName = 'Diagnostic'),
shinydashboard::menuItem(text = "Filter Cells",
tabName = "FilterCells"),
shinydashboard::menuItem(text = "Exit",
tabName = "Exit")
)
),
shinydashboard::dashboardBody(
#use shinyjs
shinyjs::useShinyjs(),
#to close window
shinyjs::extendShinyjs(text = jscode, functions = c("closeWindow")),
#use CSS
shiny::tags$head(
shiny::tags$link(rel = "stylesheet",
type = "text/css",
href = "custom.css")
),
#setting spinner
shinybusy::add_busy_spinner(
spin = "fading-circle",
position = 'bottom-right',
color = 'orange',
height = '200px',
width = '200px'
),
shinydashboard::tabItems(#home
{
shinydashboard::tabItem(
tabName = "Home",
shinydashboard::box(
width = 4,
height = '150',
title = 'Analysis name',
solidHeader = T,
background = 'black',
shiny::textInput(
inputId = 'Analysis_Name',
label = '',
value = 'Analysis'
)
),
shinydashboard::box(
width = 4,
height = '150',
title = 'Cores to use for the analysis',
solidHeader = T,
background = 'black',
align = 'center',
#cores
sliderInput(
width = '100%',
inputId = 'cores',
label = '',
value = trunc(maxCores / 2),
min = 1,
max = maxCores,
step = 1,
ticks = F
)
),
shinydashboard::box(
width = 4,
height = '150',
title = 'Output folder',
solidHeader = T,
background = 'black',
align = 'center',
shinyFiles::shinyDirButton(
id = 'Output_dir',
label = 'Output folder',
title = 'Output folder'
),
shiny::htmlOutput('Output_dir_out')
),
shinydashboard::box(
width = 4,
height = '150',
title = shiny::div(
'Chr prefix to use',
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'Depending on your reference genome, chromosomes or scaffolds could be named differently. Please, provide the prefix that precedes the chromosome or scaffold number. If no prefix is present this box has to be emptied.' , placement = 'right')
),
solidHeader = T,
background = 'black',
shiny::textInput(
inputId = 'chr_prefix',
label = '',
value = 'chr'
)
),
shiny::column(
width = 4,
height = '150',
align = 'center',
shiny::img(src = "Kronos_logo.png", height = 150)
),
shinydashboard::box(
width = 4,
height = '150',
title = shiny::div(
'Chr to include',
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'Chromosome to include in the analysis. You can define ranges using ":" and individual chromosomes using ",". Don\'t include the chr prefix!' , placement = 'right')
),
solidHeader = T,
background = 'black',
shiny::textInput(
inputId = 'chr_range',
label = '',
value = '1:22'
)
),
shinydashboard::box(
width = 4,
height = '150',
title = shiny::div(
'Chr size file',
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'For most genome assemblies this file can be downloaded from UCSC Genome Browser website.' , placement = 'right')
),
solidHeader = T,
background = 'black',
align = 'center',
shinyFiles::shinyFilesButton(
id = 'Chr_size',
label = 'Chr size file',
title = 'Chr size file',
multiple = F
),
shiny::htmlOutput('Chr_size_out', style = 'font-size: 16px;font-weight: bold;')
),
shinydashboard::box(
width = 4,
height = '150',
title = 'Final RT bin size',
solidHeader = T,
background = 'black',
align = 'center',
#cores
sliderInput(
width = '100%',
inputId = 'binsize',
label = '',
value = '0.5',
min = 0.1,
max = 2,
step = 0.05,
ticks = F,
post = 'Mb'
)
),
shinydashboard::box(
width = 4,
height = '150',
title = 'Apply settings',
solidHeader = T,
background = 'black',
align = 'center',
shinyjs::disabled(
shiny::actionButton(inputId = 'ApplySettings', label =
'Apply settings'))
)
)
},
#upload
{
shinydashboard::tabItem(
tabName = "Upload",
shinydashboard::box(
title = 'Upload samples',
solidHeader = T,
status = 'primary',
collapsible = T,
width = 12,
shiny::fluidRow(
shiny::column(
width = 3,
shinyFiles::shinyFilesButton(
id = 'PerCell_file',
label = shiny::div('PerCell file',
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = '_PerCell.csv file created by Kronos scRT processing.' , placement = 'right')
),
title = 'PerCell File',
multiple = F,
style = "width: 100%"
)
),
shiny::column(width = 9,
shiny::htmlOutput(outputId =
'PerCell_file_out'))
),
shiny::fluidRow(
shiny::column(
width = 3,
shinyFiles::shinyFilesButton(
id = 'scCN',
label = shiny::div('sc tracks file',
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = '_scCNV.tsv file created by Kronos scRT processing.' , placement = 'right')
),
title = 'sc tracks file',
multiple = F,
style = "width: 100%"
)
),
shiny::column(width = 9,
shiny::htmlOutput(outputId =
'scCN_out'))
),
shiny::fluidRow(
shiny::column(
width = 3,
shinyFiles::shinyFilesButton(
id = 'setting_file',
label = shiny::div(
'Setting file',
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'Optional. If this dataset has been used in a previous analysis the produced setting file can be reused.' , placement = 'right')
),
title = "Setting file",
multiple = F,
style = "width: 100%"
)
),
shiny::column(width = 9,
shiny::htmlOutput(outputId =
'setting_file_out'))
),
shiny::fluidRow(
shiny::column(
width = 3,
shinyFiles::shinyFilesButton(
id = 'whoiswho_file',
label = shiny::div(
"Who's who file",
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'Optional. A file containing experimental cell staging information. Such a file has to containing cell identifiers under the columns Cell and logical values under the columns S_Phase. (TRUE = a cell is in S phase, FALSE = a cell is in G1 or G2 phase)' , placement = 'right')
),
title = "Who's who file",
multiple = F,
style = "width: 100%"
)
),
shiny::column(width = 9,
shiny::htmlOutput(outputId = 'whoiswho_file_out'))
),
shiny::fluidRow(
shiny::column(
width = 6,
shiny::textInput(
inputId = ('FileName'),
value = 'Exp',
label = shiny::div(
"Basename",
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'This Name identifies each individual experiment' , placement = 'right')
),
width = '100%'
)
),
shiny::column(
width = 6,
shiny::textInput(
inputId = ('GroupName'),
value = 'Exp',
label = shiny::div(
"Group name",
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'If cells have been sequenced in different experiments, providing the same group name allows to merge them after normalization.' , placement = 'right')
),
width = '100%'
)
)
),
shiny::fluidRow(shiny::column(
width = 3,
shiny::actionButton(
inputId = 'Add_sample',
label = "Add sample",
width = '100%'
)
)),
shiny::fluidRow(shiny::column(
width = 12,
shiny::tableOutput('File_paths')
)),
shiny::fluidRow(shiny::column(
width = 3,
shiny::selectInput(
inputId = 'Remove_sample',
label = "Remove sample",
choices = 'Sample',
selected = 'Sample',
width = '100%'
)
))
),
shinydashboard::box(
title = shiny::div(
'Upload references',
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'Bulk RT references are optional. The program allows to upload maximum one reference per group.' , placement = 'right')
),
solidHeader = T,
status = 'primary',
collapsible = T,
collapsed = T,
width = 12,
shiny::fluidRow(
shiny::column(
width = 6,
shiny::selectInput(
inputId = 'reference_group',
label = 'group name',
choices = 'Select a group',
multiple = F,
selected = 'Select a group',
width = '100%'
)
),
shiny::column(
width = 6,
shiny::textInput(
inputId = 'reference_name',
value = 'Reference',
label = shiny::div(
'Reference name',
bsplus::shiny_iconlink() %>%
bsplus::bs_embed_popover(title = 'This name will be visualize on plots.' , placement = 'right')
),
width = '100%'
)
)
),
shiny::fluidRow(
shiny::column(
width = 3,
shinyFiles::shinyFilesButton(
id = 'reference_file',
label = "Reference RT file",
title = "Reference RT file",
multiple = F,
style = "width: 100%"
)
),
shiny::column(width = 9,
shiny::htmlOutput(outputId =
'reference_file_out'))
),
shiny::fluidRow(shiny::column(
width = 3,
shiny::actionButton(
inputId = 'Add_reference',
label = "Add reference",
width = '100%'
)
)),
shiny::fluidRow(shiny::column(
width = 12,
shiny::tableOutput('Reference_paths')
))
),
shiny::fluidRow(shiny::column(
width = 3,
shinyjs::disabled(
shiny::actionButton(
inputId = 'Done_upload',
label = 'Done',
width = '100%'
)
)
))
)
},
#Diagnostic
{
shinydashboard::tabItem(tabName = 'Diagnostic',
shiny::uiOutput('Diagnostic_UI'),
shiny::fluidRow(shiny::column(
width = 3,
shiny::actionButton(
inputId = 'Diagnostic_done',
label = "Start RT analysis",
width = '100%'
)
)))
},
#Filter Cells
{
shinydashboard::tabItem(
tabName = 'FilterCells',
shiny::fluidRow(shiny::column(
width = 6,
offset = 3,
shiny::sliderInput(
width = '100%',
inputId = 'min_correlation_matrix',
label = 'Min correlation to be kept',
min = 0,
max = 1,
step = 0.01,
value = 0.25
)
)),
shiny::fluidRow(
shiny::column(width = 6,
plotly::plotlyOutput('plot1_RT',width = '100%',height = 'auto')),
shiny::column(width = 6,
plotly::plotlyOutput('plot2_RT',width = '100%',height = 'auto'))
),
shiny::fluidRow(shiny::column(
width = 3,
shiny::actionButton(
inputId = 'RT_next',
label = "Finalize",
width = '100%'
)
)),
shiny::fluidRow(
shiny::column(
width = 12,
align = 'center',
shiny::htmlOutput('END')
)
)
)
})
)
)
server <- function(input, output, session) {
#variables
variables = shiny::reactiveValues(
setting_file = '',
File_paths = dplyr::tibble(),
Reference_paths = dplyr::tibble(),
PerCell_file = '',
whoiswho_file = '',
PerCell_file_error = '',
first_line_Percell = NULL,
scCN = '',
scCN_error = '',
first_line_CNV = NULL,
reference_file = '',
groups = '',
groups_with_a_reference = '',
Chr_size = '<font color=\"#FF0000\"><b> This is a mandatory input.</b></font>',
roots = c(
shinyFiles::getVolumes()(),
Home = Sys.getenv("HOME"),
OutputFolder = file.path(Sys.getenv("HOME"))
),
to_keep = NULL,
Save__scPlots = F,
Save__BinRep = F
)
#store diagnostic module info
diagnostic_module_ls = shiny::reactiveValues(ui = list(),
server = list())
#store data
data = shiny::reactiveValues(
PerCell = dplyr::tibble(),
scCN = dplyr::tibble(),
whoswho = dplyr::tibble(),
setting = dplyr::tibble(),
reference = dplyr::tibble(),
variability = dplyr::tibble(),
scRT = dplyr::tibble(),
G1G2 = dplyr::tibble()
)
#load
shiny::observe({
#add tab items in diagnostic
output$Diagnostic_UI = shiny::renderUI({
diagnostic_module_ls$ui
})
})
#stop app when the session ends
session$onSessionEnded(function() {
shiny::stopApp()
})
#home
{
shiny::observe({
shinyFiles::shinyFileChoose(
input = input,
id = 'Chr_size',
session = session,
roots = variables$roots,
defaultRoot = 'Home'
)
shinyFiles::shinyDirChoose(
input = input,
id = 'Output_dir',
session = session,
roots = variables$roots,
defaultRoot = 'Home'
)
shiny::observeEvent(input$Analysis_Name, {
updateTextInput(
session = session,
inputId = 'Analysis_Name',
value = stringr::str_replace_all(
string = input$Analysis_Name,
pattern = ' ',
replacement = '_'
)
)
})
output$Output_dir_out <-
renderText(paste(
'<H4><b>',
file.path(variables$roots['OutputFolder'], input$Analysis_Name),
'</H4></b>'
))
if (!stringr::str_detect(string = variables$Chr_size,
pattern = '<font color=\"#FF0000\">')) {
shinyjs::enable('Upload_chr_size')
} else{
shinyjs::disable('Upload_chr_size')
}
})
#load chrs_size
shiny::observeEvent(input$Chr_size, {
if (!is.numeric(input$Chr_size)) {
#get path
variables$Chr_size = dplyr::tibble(
shinyFiles::parseFilePaths(
roots = variables$roots ,
selection = input$Chr_size
)
) %>% dplyr::pull(datapath)
variables$Chr_size = Kronos.scRT::right_format(
file_path = variables$Chr_size,
delim = '\t',
columns_to_check = 2,
wrong_message = paste0(
'<b>',
variables$Chr_size,
'<font color=\"#FF0000\"> does not have the right format. </b></font>'
),
rigth_message = variables$Chr_size
)
} else{
variables$Chr_size = '<font color=\"#FF0000\"><b> This is a mandatory input.</b></font>'
}
})
shiny::observe(output$Chr_size_out <-
renderText(variables$Chr_size))
shiny::observeEvent(input$Output_dir, {
if (!is.numeric(input$Output_dir)) {
variables$roots = c(
shinyFiles::getVolumes()(),
Home = Sys.getenv("HOME"),
OutputFolder = parseDirPath(
roots = variables$roots,
selection = input$Output_dir
)
)
}
})
shiny::observeEvent(input$Chr_size,{
if (!is.numeric(input$Chr_size) & input$ApplySettings == 0) {
shinyjs::enable('ApplySettings')
} else{
shinyjs::disable('ApplySettings')
}
})
shiny::observeEvent(input$ApplySettings, {
if (input$ApplySettings == 1) {
shinyjs::disable('ApplySettings')
shinyjs::disable('binsize')
shinyjs::disable('Chr_size')
shinyjs::disable('chr_range')
shinyjs::disable('chr_prefix')
shinyjs::disable('Output_dir')
shinyjs::disable('cores')
shinyjs::disable('Analysis_Name')
#move to next tab
shinydashboard::updateTabItems(session = session,
inputId = 'Sidebar',
selected = 'Upload')
}
})
}
#load data
{
#samples
{
shiny::observe({
shinyFiles::shinyFileChoose(
input = input,
id = 'PerCell_file',
roots = variables$roots,
session = session,
defaultRoot = 'Home'
)
shinyFiles::shinyFileChoose(
input = input,
id = 'scCN',
roots = variables$roots,
session = session,
defaultRoot = 'Home'
)
shinyFiles::shinyFileChoose(
input = input,
id = 'whoiswho_file',
roots = variables$roots,
session = session,
defaultRoot = 'Home'
)
shinyFiles::shinyFileChoose(
input = input,
id = 'setting_file',
roots = variables$roots,
session = session,
defaultRoot = 'Home'
)
})
#activate deactivate Add input button
shiny::observe({
if (variables$PerCell_file == '' | variables$scCN == '' |
stringr::str_detect(string = variables$PerCell_file, pattern = ' does not have the right format') |
stringr::str_detect(string = variables$scCN, pattern = ' does not have the right format') |
stringr::str_detect(string = variables$whoiswho_file, pattern = ' does not have the right format') |
input$ApplySettings != 1) {
shinyjs::disable(id = 'Add_sample')
output$PerCell_file_out = shiny::renderText(variables$PerCell_file)
output$scCN_out = shiny::renderText(variables$scCN)
} else{
if (all(!c('basename', 'group') %in% names(variables$first_line_Percell)) &
all(!c('basename', 'group') %in% names(variables$first_line_CNV))) {
#rest outputs if they had changed
output$PerCell_file_out = shiny::renderText(variables$PerCell_file)
output$scCN_out = shiny::renderText(variables$scCN)
shinyjs::enable(id = 'Add_sample')
} else if (all(c('basename', 'group') %in% names(variables$first_line_Percell)) &
all(c('basename', 'group') %in% names(variables$first_line_CNV))) {
if (variables$first_line_Percell$basename == variables$first_line_CNV$basename &
variables$first_line_Percell$group == variables$first_line_CNV$group) {
#update file name and group name if they match
updateTextInput(
session = session,
inputId = 'FileName',
value = variables$first_line_Percell$basename
)
updateTextInput(
session = session,
inputId = 'GroupName',
value = variables$first_line_Percell$group
)
#rest outputs if they had changed
output$PerCell_file_out = shiny::renderText(variables$PerCell_file)
output$scCN_out = shiny::renderText(variables$scCN)
shinyjs::enable(id = 'Add_sample')
} else{
#set warning
output$PerCell_file_out = shiny::renderText(
'<font color=\"#FF0000\"> Basename and Group do not match between PerCell and sc tracks files!!! </b></font>'
)
output$scCN_out = shiny::renderText(
'<font color=\"#FF0000\"> Basename and Group do not match between PerCell and sc tracks files!!! </b></font>'
)
shinyjs::disable(id = 'Add_sample')
}
} else{
output$PerCell_file_out = shiny::renderText(
'<font color=\"#FF0000\"> Basename and Group do not match between PerCell and sc tracks files!!! </b></font>'
)
output$scCN_out = shiny::renderText(
'<font color=\"#FF0000\"> Basename and Group do not match between PerCell and sc tracks files!!! </b></font>'
)
shinyjs::disable(id = 'Add_sample')
}
}
#activate button to srat RT analysis
if (nrow(variables$File_paths) > 0) {
shinyjs::enable(id = 'Diagnostic_done')
shinyjs::enable(id = 'Done_upload')
} else{
shinyjs::disable(id = 'Diagnostic_done')
shinyjs::disable(id = 'Done_upload')
}
})
#move to next page
shiny::observeEvent(input$Done_upload, {
shinydashboard::updateTabItems(session = session,
inputId = 'Sidebar',
selected = 'Diagnostic')
})
#replace ' ' with '_'
shiny::observeEvent(input$FileName, {
updateTextInput(
session = session,
inputId = 'FileName',
value = stringr::str_replace_all(
string = input$FileName,
pattern = ' ',
replacement = '_'
)
)
})
shiny::observeEvent(input$GroupName, {
updateTextInput(
session = session,
inputId = 'GroupName',
value = stringr::str_replace_all(
string = input$GroupName,
pattern = ' ',
replacement = '_'
)
)
})
# get PerCell_file path and check if it has the right format
shiny::observeEvent(input$PerCell_file, {
if (!is.numeric(input$PerCell_file)) {
#get path
variables$PerCell_file = dplyr::tibble(
shinyFiles::parseFilePaths(
roots = variables$roots ,
selection = input$PerCell_file
)
) %>% dplyr::pull(datapath)
#check if it is the right file
variables$PerCell_file = Kronos.scRT::right_format(
file_path = variables$PerCell_file,
delim = ',',
columns_to_check = c(
"Cell",
"normalized_dimapd",
"mean_ploidy",
"ploidy_confidence",
"is_high_dimapd",
"is_noisy",
"coverage_per_1Mbp"
),
wrong_message = paste0(
'<b>',
variables$PerCell_file,
'<font color=\"#FF0000\"> does not have the right format. </b></font>'
),
rigth_message = variables$PerCell_file
)
#read line one and look for group and basename
if (stringr::str_detect(
string = variables$PerCell_file,
pattern = ' does not have the right format',
negate = T
)) {
variables$first_line_Percell = readr::read_csv(variables$PerCell_file, n_max = 1)
}
} else{
variables$PerCell_file = ''
}
})
# get scCN path and check if it has the right format
shiny::observeEvent(input$scCN, {
if (!is.numeric(input$scCN)) {
#get path
variables$scCN = dplyr::tibble(
shinyFiles::parseFilePaths(roots = variables$roots ,
selection = input$scCN)
) %>% dplyr::pull(datapath)
#check if it is the right file
variables$scCN = Kronos.scRT::right_format(
file_path = variables$scCN,
delim = '\t',
columns_to_check = c('chr',
'start',
'end' ,
'copy_number',
'reads' ,
'Cell'),
wrong_message = paste0(
'<b>',
variables$scCN,
'<font color=\"#FF0000\"> does not have the right format. </b></font>'
),
rigth_message = variables$scCN
)
#read line one and look for group and basename
if (stringr::str_detect(
string = variables$scCN,
pattern = ' does not have the right format',
negate = T
)) {
variables$first_line_CNV = readr::read_tsv(variables$scCN, n_max = 1)
}
} else{
variables$scCN = ''
}
})
# get whoiswho path and check if it has the right format
shiny::observeEvent(input$whoiswho_file, {
if (!is.numeric(input$whoiswho_file)) {
#get path
variables$whoiswho_file = dplyr::tibble(
shinyFiles::parseFilePaths(
roots = variables$roots ,
selection = input$whoiswho_file
)
) %>% dplyr::pull(datapath)
#check if it is the right file
variables$whoiswho_file = Kronos.scRT::right_format(
file_path = variables$whoiswho_file,
delim = '\t',
columns_to_check = c('Cell', 'Phase'),
wrong_message = paste0(
'<b>',
variables$whoiswho_file,
'<font color=\"#FF0000\"> does not have the right format. </b></font>'
),
rigth_message = variables$whoiswho_file
)
output$whoiswho_file_out = shiny::renderText(variables$whoiswho_file)
} else{
variables$whoiswho_file = ''
output$whoiswho_file_out = shiny::renderText(variables$whoiswho_file)
}
})
# get setting path and check if it has the right format
shiny::observeEvent(input$setting_file, {
if (!is.numeric(input$setting_file)) {
#get path
variables$setting_file = dplyr::tibble(
shinyFiles::parseFilePaths(
roots = variables$roots ,
selection = input$setting_file
)
) %>% dplyr::pull(datapath)
#check if it is the right file
variables$setting_file = Kronos.scRT::right_format(
file_path = variables$setting_file,
delim = '\t',
columns_to_check = c(
'threshold_Sphase' ,
'threshold_G1G2phase' ,
'Sphase_first_part',
'Sphase_second_part' ,
'RPMPH_TH',
'RPM_TH'
),
wrong_message = paste0(
'<b>',
variables$setting_file,
'<font color=\"#FF0000\"> does not have the right format. </b></font>'
),
rigth_message = variables$setting_file
)
output$setting_file_out = shiny::renderText(variables$setting_file)
} else{
variables$setting_file = ''
output$setting_file_out = shiny::renderText(variables$setting_file)
}
})
#add samples to dataframe
shiny::observeEvent(input$Add_sample, {
shinyjs::disable(id = 'Add_sample')
shinyjs::disable(id = 'Remove_sample')
#create directory if it does not exist
if (!dir.exists(file.path(variables$roots['OutputFolder'], input$Analysis_Name))) {
dir.create(file.path(variables$roots['OutputFolder'], input$Analysis_Name),
recursive = T)
}
#check if it is a new group
if (!(
input$GroupName %in% variables$groups |
input$GroupName %in% variables$groups_with_a_reference
)) {
variables$groups = c(variables$groups, input$GroupName)
}
#new line
newline = dplyr::tibble(
PerCellFile = basename(variables$PerCell_file),
scCNV = basename(variables$scCN),
basename = input$FileName,
group = input$GroupName,
WhoSWho = ifelse(
!is.numeric(input$whoiswho_file),
basename(variables$whoiswho_file),
''
),
Setting = ifelse(
!is.numeric(input$setting_file),
basename(variables$setting_file),
''
)
)
#load data
#if whoiswho_file's been provided
if (!is.numeric(input$whoiswho_file)) {
data$PerCell = rbind(
data$PerCell,
readr::read_csv(variables$PerCell_file, col_types = 'cnnnlln') %>%
dplyr::mutate(
basename = input$FileName,
group = input$GroupName
)
) %>%
dplyr::inner_join(readr::read_tsv(variables$whoiswho_file, col_types = 'cc'),
by = c("Cell")) %>%
dplyr::mutate(
is_high_dimapd = ifelse(Phase == 'S', T, F),
is_noisy = ifelse(is_high_dimapd, T, is_noisy)
) %>%
dplyr::select(-Phase)
} else{
data$PerCell = rbind(
data$PerCell,
readr::read_csv(variables$PerCell_file, col_types = 'cnnnlln') %>%
dplyr::mutate(
basename = input$FileName,
group = input$GroupName
)
)
}
data$scCN = rbind(
data$scCN,
readr::read_tsv(variables$scCN, col_types = 'ccnnnn') %>%
dplyr::mutate(
basename = input$FileName,
group = input$GroupName
)
)
if (!is.numeric(input$whoiswho_file) &
variables$whoiswho_file != '') {
data$whoswho = rbind(
data$whoswho,
readr::read_tsv(variables$whoiswho_file, col_types = 'cc') %>%
dplyr::mutate(
basename = input$FileName,
group = input$GroupName
)
)
}
#if setting_file's been provided
if (!is.numeric(input$setting_file) &
variables$setting_file != '') {
data$setting = rbind(
data$setting,
readr::read_tsv(variables$setting_file, col_types = 'nnnnnnncc') %>%
dplyr::mutate(
basename = input$FileName,
group = input$GroupName
)
)
}
#add line to dataframe
variables$File_paths = rbind(variables$File_paths,
newline)
#call diagnostic module
diagnostic_module_ls$ui[[paste0(input$FileName)]] = diagnostic_ui(id = input$FileName)
diagnostic_module_ls$server[[paste0(input$FileName)]] = diagnostic_server(
id = input$FileName,
PerCell = data$PerCell %>%
dplyr::filter(basename ==
input$FileName),
#if setting is not there return empty tibble
Setting = tryCatch(
data$setting %>%
dplyr::filter(basename ==
input$FileName),
error =
function(x)
return(dplyr::tibble())
),
Cores = input$cores,
OutFolder = file.path(variables$roots['OutputFolder'], input$Analysis_Name)
)
#update Remove_sample
shiny::updateSelectInput(
inputId = 'Remove_sample',
choices = c('Sample', variables$File_paths$basename)
)
# reset inputs
shiny::updateTextInput(
session = session,
inputId = 'FileName',
value = paste('Exp', input$Add_sample, sep = '_')
)
shiny::updateTextInput(
session = session,
inputId = 'GroupName',
value = paste('Exp', input$Add_sample, sep = '_')
)
variables$PerCell_file = ''
variables$scCN = ''
variables$whoiswho_file = ''
variables$setting_file = ''
output$scCN_out = shiny::renderText('')
output$PerCell_file_out = shiny::renderText('')
output$whoiswho_file_out = shiny::renderText('')
output$setting_file_out = shiny::renderText('')
output$File_paths = shiny::renderTable(variables$File_paths)
#update groups for references
shiny::updateSelectInput(
inputId = 'reference_group',
choices = c('Select a group',
variables$groups),
selected = 'Select a group'
)
#reactivate remove sample
shinyjs::enable('Remove_sample')
})
#remove samples
shiny::observeEvent(input$Remove_sample, {
if (input$Remove_sample != 'Sample') {
#remove
variables$File_paths = variables$File_paths %>%
dplyr::filter(basename != input$Remove_sample)
# stop module
diagnostic_module_ls$ui[[input$Remove_sample]] = NULL
diagnostic_module_ls$server[[input$Remove_sample]] = NULL
#update Remove_sample
shiny::updateSelectInput(
inputId = 'Remove_sample',
choices = c('Sample', variables$File_paths$basename),
selected = 'Sample'
)
#update groups for Reference
variables$groups = variables$groups[variables$groups %in% unique(variables$File_paths$group)]
shiny::updateSelectInput(
inputId = 'reference_group',
choices = c('Select a group',
variables$groups),
selected = 'Select a group'
)
}
})
shiny::observeEvent(input$Diagnostic_done, {
#disable button
shinyjs::disable('Diagnostic_done')
#find all the setting files created by diagnostic
files_to_load = list.files(
path = file.path(variables$roots['OutputFolder'], input$Analysis_Name),
pattern = '_settings.txt',
full.names = T
)
# select those with the right format
files_to_load = files_to_load[unlist(
lapply(
files_to_load,
Kronos.scRT::right_format,
delim = '\t',
columns_to_check = c(
'threshold_Sphase' ,
'threshold_G1G2phase' ,
'Sphase_first_part',
'Sphase_second_part' ,
'Ploidy',
'RPMPH_TH',
'RPM_TH',
'basename',
'group'
),
wrong_message = F,
rigth_message = T
)
)]
#upload new setting file
data$setting = Kronos.scRT::load_multiple_df(dirs = files_to_load,
delim = '\t',
col_types = 'nnnnnnncc')
########RT Start first part
{
variables$chr_list = paste0(input$chr_prefix,
unlist(
Kronos.scRT::String_to_Range(x = stringr::str_split(input$chr_range, ',')[[1]])
))
#upload chr sizes
data$Chr_size = readr::read_tsv(
variables$Chr_size,
col_names = c('chr', 'size'),
col_types = 'cn'
) %>%
dplyr::filter(chr %in% variables$chr_list) %>%
dplyr::mutate(chr = factor(x = chr, levels = variables$chr_list)) %>%
tidyr::drop_na()
variables$chr_list = variables$chr_list[variables$chr_list %in% unique(data$Chr_size$chr)]
#Apply Settings to PerCell
data$PerCell = Kronos.scRT::AdjustPerCell(
PerCell = data$PerCell ,
Settings = data$setting,
Basename_leves = variables$File_paths$basename ,
Group_leves = unique(variables$File_paths$group)
)
#Correct scCN based on PerCell
data$scCN = Kronos.scRT::AdjustCN(
scCN = data$scCN,
PerCell = data$PerCell ,
Basename_leves = variables$File_paths$basename ,
Group_leves = unique(variables$File_paths$group),
Chr_filter = variables$chr_list
)
#create bins
bins = Kronos.scRT::GenomeBinning(
Chr_size = data$Chr_size,
size = input$binsize * 10 ^ 6,
Chr_filter = variables$chr_list,
Cores = input$cores
)
#rebin data
data$signal_smoothed = Kronos.scRT::Rebin(
PerCell = data$PerCell,
scCN = data$scCN,
Bins = bins,
Sphase = T
)
G1G2_smoothed = Kronos.scRT::Rebin(
PerCell = data$PerCell,
scCN = data$scCN,
Bins = bins,
Sphase = F
)
#calculate background
median_G1G2_profile = Kronos.scRT::BackGround(G1_scCN = G1G2_smoothed)
# create single G1/G2 single cell file
G1G2_smoothed = Kronos.scRT::Replication_state(
Samples = G1G2_smoothed,
background = median_G1G2_profile,
Chr_filter = variables$chr_list,
cores = input$cores
)
#reshape and save
G1G2_smoothed = G1G2_smoothed %>%
dplyr::select(
chr,
start,
end,
CN,
background,
CN_bg,
th,
Rep,
PercentageReplication,
Cell,
basename,
group,
newIndex
)
G1G2_smoothed %>%
readr::write_tsv(
file = file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
paste0(
input$Analysis_Name,
'_G1_G2_single_cells_CNV_',
input$binsize,
'Mb.tsv'
)
),
col_names = T
)
data$G1G2 = G1G2_smoothed
if (nrow(data$reference) != 0) {
# rebin reference RT
data$reference = Kronos.scRT::RebinRT(
RT = data$reference,
Bins = bins,
Chr_filter = variables$chr_list
)
#write output
readr::write_tsv(
x = data$reference,
file = file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
paste0(
input$Analysis_Name,
'_reference_replication_timing_',
input$binsize,
'Mb.tsv'
)
),
col_names = T
)
}
#Calculate replication state S
data$signal_smoothed = Kronos.scRT::Replication_state(
Samples = data$signal_smoothed,
background = median_G1G2_profile,
Chr_filter = variables$chr_list,
cores = input$cores
)
# remove control track
rm('median_G1G2_profile')
#matrix for the correlation
data$signal_smoothed = data$signal_smoothed %>%
dplyr::ungroup() %>%
dplyr::arrange(group, newIndex) %>%
tidyr::unite(index, c(group, newIndex), sep = ' _ ') %>%
dplyr::mutate(index = factor(index, levels = unique(index)))
mat = data$signal_smoothed %>%
tidyr::unite(pos, c(chr, start), sep = ':') %>%
dplyr::mutate(Rep = as.numeric(Rep)) %>%
dplyr::select(pos, index, Rep) %>%
tidyr::spread(key = index, value = Rep) %>%
tibble::column_to_rownames('pos') %>%
dplyr::filter(complete.cases(.)) %>%
as.matrix()
#index
variables$Index = colnames(mat)
#correlation similarity distance
variables$results = 1 - as.matrix(ade4::dist.binary(
t(mat),
method = 2,
diag = T,
upper = T
))
variables$basenames = dplyr::tibble(Group = stringr::str_remove(colnames(mat), ' _ [0-9]{1,10}$'))
#write matrix and plot heatmap before filtering
saveRDS(
object = variables$results,
file = file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
paste0(
input$Analysis_Name,
'_correlation_per_cell_before_filtering.rds'
)
)
)
#prepare color patterns
variables$color = grDevices::colorRampPalette(
colors = c(
"#00204DFF",
"#233E6CFF",
"#575C6DFF",
"#7C7B78FF",
"#A69D75FF",
"#D3C164FF",
"#FFEA46FF"
)
)
output$plot1_RT <- plotly::renderPlotly(
{heatmaply::heatmaply(
x = variables$results,
colors = variables$color,
dendrogram = F,
showticklabels = F,
row_side_colors = variables$basenames,
col_side_colors = variables$basenames,
limits = c(0, 1)
) %>%
plotly::layout(
showlegend = FALSE,
legend = FALSE,
annotations = list(visible = FALSE)
)}
)
variables$to_keep = foreach::foreach(i = unique(variables$basenames$Group)) %do% {
sub_mat = variables$results[variables$basenames$Group == i, variables$basenames$Group == i]
diag(sub_mat) = 0
! matrixStats::rowQuantiles(x = sub_mat,
probs = 0.60,
na.rm = T) <= input$min_correlation_matrix
}
variables$to_keep = unlist(variables$to_keep)
variables$results_after_filtering = variables$results[variables$to_keep, variables$to_keep]
variables$basenames_after_filtering = variables$basenames[variables$to_keep,]
output$plot2_RT <- plotly::renderPlotly({
heatmaply::heatmaply(
x = variables$results_after_filtering,
colors = variables$color,
dendrogram = F,
showticklabels = F,
row_side_colors = variables$basenames_after_filtering,
col_side_colors = variables$basenames_after_filtering,
limits = c(0, 1)
) %>%
plotly::layout(
showlegend = FALSE,
legend = FALSE,
annotations = list(visible = FALSE)
)}
)
#bring the user to FilterCells
shinydashboard::updateTabItems(session = session,
inputId = 'Sidebar',
selected = 'FilterCells')
}
########RT end first part
})
###filter
shiny::observeEvent(input$min_correlation_matrix, {
if (!is.null(variables$to_keep)) {
variables$to_keep = foreach::foreach(i = unique(variables$basenames$Group)) %do% {
sub_mat = variables$results[variables$basenames$Group == i, variables$basenames$Group == i]
diag(sub_mat) = 0
! matrixStats::rowQuantiles(x = sub_mat,
probs = 0.60,
na.rm = T) <= input$min_correlation_matrix
}
variables$to_keep = unlist(variables$to_keep)
variables$results_after_filtering = variables$results[variables$to_keep, variables$to_keep]
variables$basenames_after_filtering = variables$basenames[variables$to_keep,]
output$plot2_RT <- plotly::renderPlotly(
heatmaply::heatmaply(
x = variables$results_after_filtering,
colors = variables$color,
dendrogram = F,
showticklabels = F,
row_side_colors = variables$basenames_after_filtering,
col_side_colors = variables$basenames_after_filtering,
limits = c(0, 1)
) %>%
plotly::layout(
showlegend = FALSE,
legend = FALSE,
annotations = list(visible = FALSE)
)
)
}
})
########RT second part start
shiny::observeEvent(input$RT_next, {
if (input$RT_next == 1) {
shinyjs::disable('RT_next')
#write matrix and plot heatmap after filtering
saveRDS(
object = variables$results_after_filtering,
file = file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
paste0(
input$Analysis_Name,
'_correlation_per_cell_after_filtering.rds'
)
)
)
variables$Index = variables$Index[variables$to_keep]
#filter out samples that don't correlate and save
data$signal_smoothed = data$signal_smoothed %>%
dplyr::filter(index %in% variables$Index) %>%
tidyr::separate(index, c('group', 'index'), sep = ' _ ') %>%
dplyr::mutate(group = factor(group, level = unique(variables$File_paths$group)))
rep_percentage = data$signal_smoothed %>%
dplyr::group_by(Cell, basename, group, index) %>%
dplyr::summarise(Rep_percentage = mean(Rep))
#new index
new_index_list = rep_percentage %>%
dplyr::ungroup() %>%
dplyr::arrange(Rep_percentage) %>%
dplyr::group_by(group) %>%
dplyr::mutate(newIndex = 1:dplyr::n()) %>%
dplyr::arrange(group, newIndex) %>%
dplyr::select(oldIndex = index, newIndex, Cell, basename, group)
data$signal_smoothed = data$signal_smoothed %>%
dplyr::ungroup() %>%
dplyr::inner_join(new_index_list,
by = c('Cell', 'index' = 'oldIndex', 'basename', 'group')) %>%
dplyr::select(
chr,
start,
end,
CN,
background,
CN_bg,
th,
Rep,
PercentageReplication,
Cell,
basename,
group,
newIndex
)
readr::write_tsv(
x = data$signal_smoothed,
file = file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
paste0(
input$Analysis_Name,
'_single_cells_CNV_',
input$binsize,
'Mb.tsv'
)
),
col_names = T
)
#select used data and save the new per cell files
used_cells = rbind(
new_index_list %>%
dplyr::select(Cell, basename, group) %>%
dplyr::ungroup(),
data$PerCell %>%
dplyr::filter(Type == 'G1/G2-phase cells') %>%
dplyr::select(Cell, basename, group) %>%
dplyr::ungroup()
)
data$PerCell = data$PerCell %>%
dplyr::inner_join(used_cells,
Joining,
by = c("Cell", "basename", "group"))
#create folder
if (!dir.exists(
file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
'Cells_used_in_the_analysis_info'
)
)) {
dir.create(
file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
'Cells_used_in_the_analysis_info'
),
recursive = T
)
}
#isolate for foreach
PerCell = shiny::isolate(data$PerCell)
bs = foreach::foreach(i = unique(PerCell$basename)) %do% {
PerCell %>%
dplyr::filter(basename == i) %>%
dplyr::select(
Cell,
normalized_dimapd,
mean_ploidy,
ploidy_confidence,
is_high_dimapd,
is_noisy,
coverage_per_1Mbp
) %>%
readr::write_csv(
file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
'Cells_used_in_the_analysis_info',
paste0(
input$Analysis_Name,
i,
'_per_Cell_summary_metrics.csv'
)
)
)
i
}
rm('bs')
rm('PerCell')
rm('new_index_list')
#calculate psudo bulk
data$scRT = Kronos.scRT::pseudoBulkRT(data$signal_smoothed)
readr::write_tsv(
x = data$scRT,
file = file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
paste0(
input$Analysis_Name,
'_calculated_replication_timing_',
input$binsize,
'Mb.tsv'
)
),
col_names = T
)
#merge reference and scRT
data$RTs = rbind(data$scRT %>% dplyr::ungroup(),
data$reference %>% dplyr::ungroup())
#prepare for twidth
data$variability = Kronos.scRT::Variability(S_scCN = data$signal_smoothed,
scRT = data$scRT)
readr::write_tsv(
x = data$variability,
file = file.path(
variables$roots['OutputFolder'],
input$Analysis_Name,
paste0(input$Analysis_Name,
'_scRT_variability.tsv')
),
col_names = T
)
output$END <- shiny::renderText({'<h3><font color=\"#FF0000\"> DONE! </b></font></h3>' })
}
})
########RT second part end
}
#reference
{
shiny::observe({
shinyFiles::shinyFileChoose(
input = input,
id = 'reference_file',
roots = variables$roots,
session = session,
defaultRoot = 'Home'
)
if (file.exists(variables$reference_file) &
input$reference_group != 'Select a group') {
shinyjs::enable('Add_reference')
} else{
shinyjs::disable('Add_reference')
}
})
#select and check reference file
shiny::observeEvent(input$reference_file, {
if (!is.numeric(input$reference_file)) {
#get path
variables$reference_file = dplyr::tibble(
shinyFiles::parseFilePaths(
roots = variables$roots ,
selection = input$reference_file
)
) %>% dplyr::pull(datapath)
#check if it is the right file
variables$reference_file = Kronos.scRT::right_format(
file_path = variables$reference_file,
delim = '\t',
columns_to_check = 4,
wrong_message = paste0(
'<b>',
variables$reference_file,
'<font color=\"#FF0000\"> does not have the right format. </b></font>'
),
rigth_message = variables$reference_file
)
output$reference_file_out = shiny::renderText(variables$reference_file)
} else{
variables$reference_file = ''
output$reference_file_out = shiny::renderText(variables$reference_file)
}
})
#replace ' ' with '_' for reference name
shiny::observeEvent(input$reference_name, {
updateTextInput(
session = session,
inputId = 'reference_name',
value = stringr::str_replace_all(
string = input$reference_name,
pattern = ' ',
replacement = '_'
)
)
})
#add reference
shiny::observeEvent(input$Add_reference, {
shinyjs::disable('Add_reference')
data$reference = rbind(
data$reference,
readr::read_tsv(
variables$reference_file,
col_names = c('chr', 'start', 'end', 'RT'),
col_types = 'cnnn'
) %>%
dplyr::mutate(
basename = input$reference_name,
group = input$reference_group
)
)
variables$Reference_paths = rbind(
variables$Reference_paths,
dplyr::tibble(
File = basename(variables$reference_file),
basename = input$reference_name,
group = input$reference_group
)
)
output$Reference_paths = shiny::renderTable(variables$Reference_paths)
variables$reference_file = ''
output$reference_file_out = shiny::renderText('')
variables$groups_with_a_reference = c(variables$groups_with_a_reference,
input$reference_group)
variables$groups = variables$groups[!variables$groups %in% variables$groups_with_a_reference]
#update groups for Reference
shiny::updateSelectInput(
inputId = 'reference_group',
choices = c('Select a group',
variables$groups),
selected = 'Select a group'
)
updateTextInput(inputId = 'reference_name',
value = paste0('Reference_', input$Add_reference))
})
}
}
####exit
{
shiny::observe({
if (input$Sidebar == 'Exit') {
shinyjs::js$closeWindow()
shiny::stopApp()
}
})
}
}
# Run the application
shiny::shinyApp(ui = ui,
server = server)
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.