In ningzhibin/rmdocpu: Metalab report upon provided templates

Readme

• The following scripts show how to use r to upload a metalab result txt file to the opencpu server as the input, options with a meta file, to render an html report then download the report to local storage
• This function could be done by any script language, such as java, perl or bash
• By doing this, Metalab can have the report easily, without the need to setup the r environment, but easily get all the report, especially with Various data visualization in the report

• These report are all html based and could be linked/iframed into a another frame file

• This doc shows the detailed method about how to test and debug the templates locally and remotely if this package is installed on remote server

• Current server is on compute Canada. The server is configured with a open cpu web server and open to the world.

• About how to make opencpu request is detailed here: https://www.opencpu.org/api.html

Required Pacakges

#library(jsonlite)
#library(httr)

preparation

• prepare the 5 result files from metalab from the same data set, and meta information table for this experiment, with the name of metainfo.txt.

see table format requirement for more information: https://github.com/northomics/Metalab_development_wiki/wiki

• prepare the 5 markdown files for local test
• test locally first then remotely

• open the correstpoding rmarkdown file to debug

• setup the working directory as this file

summary.txt

local test

without meta

data_table <- rio::import("./extdata/summary.txt", header = TRUE,check.names = FALSE, stringsAsFactors = FALSE)
# knit to all supported format (here we support pdf and html)
rmarkdown::render("./rmd/MQ_report_summary.Rmd",output_format = "html_document", params = list(data_table =  data_table), output_dir = getwd())

with meta

data_table <- rio::import("./extdata/summary.txt", header = TRUE,check.names = FALSE, stringsAsFactors = FALSE)
meta_table <- rio::import("./extdata/metainfo.txt", header = TRUE,check.names = FALSE, stringsAsFactors = FALSE) 

rmarkdown::render("./rmd/MQ_report_summary.Rmd",output_format = "html_document", params = list(data_table =  data_table, meta_table = meta_table),  output_dir = getwd())

remote test

url_api <- "http://206.12.91.148/ocpu/library/rmdocpu/R/render_MQsummary_file"

# get the root url
url_api_split <- strsplit(url_api, "/")[[1]]
url_server<- paste0(url_api_split[1],"//", url_api_split[3],"/")

without meta

# upload file and do the rendering
# in this case, the summary.txt is in the working dir. it can be anywhere with the path
# meta file is optional

r <- httr::POST(url_api, body = list(file = httr::upload_file("summary.txt")))
#r <- httr::POST(url_api, body = list(file = httr::upload_file("summary.txt"), meta = httr::upload_file("summary_metainfo.txt")))

# get all the paths of all files from the opencpu end, and locate the one, which is the report
# this step needs to be done in the script enviroment

paths <- strsplit(rawToChar(r$content), "\n")[[1]]
path_target <- paths[grep("output.html",paths)]

# save/download the report file to local storage
# the file  "maxquant_result_summary.html" now is the report
curl::curl_download(paste0(url_server, path_target), "report_ID_summary.html")

with meta

# upload file and do the rendering
# in this case, the summary.txt is in the working dir. it can be anywhere with the path
# meta file is optional

#r <- httr::POST(url_api, body = list(file = httr::upload_file("summary.txt")))
r <- httr::POST(url_api, body = list(file = httr::upload_file("summary.txt"), meta = httr::upload_file("metainfo.txt")))

# get all the paths of all files from the opencpu end, and locate the one, which is the report
# this step needs to be done in the script enviroment

paths <- strsplit(rawToChar(r$content), "\n")[[1]]
path_target <- paths[grep("output.html",paths)]

# save/download the report file to local storage
# the file  "maxquant_result_summary.html" now is the report
curl::curl_download(paste0(url_server, path_target), "report_ID_summary.html")

proteingroups.txt

• proteinGroups_report.txt is a shorted and simplified version of proteinGroups.txt, much smaller in size, more suitable for report
• proteinGroups_report.txt does not have the correct number of some items, e.g. the number of contaminant etc

local test

without meta

data_table  <- read.delim("proteinGroups_report.txt", header = TRUE,check.names = FALSE, stringsAsFactors = FALSE) # NOTE the read in options

rmarkdown::render("MQ_report_proteinGroups.Rmd",output_format = "html_document", params = list(input_datatable =  data_table), output_file="report_proteinGroups_summary.html")

with meta

data_table  <- read.delim("proteinGroups.txt", header = TRUE,check.names = FALSE, stringsAsFactors = FALSE) # NOTE the read in options
meta_table <- read.delim("metainfo.txt", header = TRUE, check.names = FALSE, stringsAsFactors = FALSE) # with meta file


rmarkdown::render("MQ_report_proteinGroups.Rmd",output_format = "html_document", params = list(input_datatable =  data_table, meta_table = meta_table), output_file="report_proteinGroups_summary.html")

remote test

without meta

url_api <- "http://206.12.91.148/ocpu/library/rmdocpu/R/render_proteinGroups_file"

# get the root url
url_api_split <- strsplit(url_api, "/")[[1]]
url_server<- paste0(url_api_split[1],"//", url_api_split[3],"/")


# upload file and do the rendering
# in this case, the proteinGroups.txt is in the working dir. it can be anywhere with the path
# variable r is the returning information from the curl function
r <- httr::POST(url_api, body = list(file = httr::upload_file("proteinGroups_report.txt")))
#r <- httr::POST(url_api, body = list(file = httr::upload_file("proteinGroups1.txt"), meta = httr::upload_file("proteinGroups1_meta.txt")), httr::timeout(200000))

# get all the paths of all files from the opencpu end, and locate the one, which is the report
# this step needs to be done in the script enviroment

paths <- strsplit(rawToChar(r$content), "\n")[[1]]
path_target <- paths[grep("output.html",paths)]
#paths

# save/download the report file to local storage
# the file  "maxquant_result_summary.html" now is the report
curl::curl_download(paste0(url_server, path_target), "report_proteinGroups_summary.html")

with meta

url_api <- "http://206.12.91.148/ocpu/library/rmdocpu/R/render_proteinGroups_file"

# get the root url
url_api_split <- strsplit(url_api, "/")[[1]]
url_server<- paste0(url_api_split[1],"//", url_api_split[3],"/")


# upload file and do the rendering
# in this case, the proteinGroups.txt is in the working dir. it can be anywhere with the path
# variable r is the returning information from the curl function
#r <- httr::POST(url_api, body = list(file = httr::upload_file("final_proteins.tsv")))
r <- httr::POST(url_api, body = list(file = httr::upload_file("proteinGroups_report.txt"), meta = httr::upload_file("metainfo.txt")), httr::timeout(200000))


# get all the paths of all files from the opencpu end, and locate the one, which is the report
# this step needs to be done in the script enviroment

paths <- strsplit(rawToChar(r$content), "\n")[[1]]
path_target <- paths[grep("output.html",paths)]
paths


# save/download the report file to local storage
# the file  "maxquant_result_summary.html" now is the report
curl::curl_download(paste0(url_server, path_target), "report_proteinGroups_summary.html")

peptide.txt test

locally

without meta

# test on locoal drive, using local rmd
data_table <- rio::import("./extdata/peptides_report.txt", header = TRUE,check.names = FALSE, stringsAsFactors = FALSE)

rmarkdown::render("./rmd/MQ_report_peptides.Rmd",output_format = "html_document", params = list(data_table =  data_table), output_dir = getwd())

with meta

data_table  <- rio::import("./extdata/peptides_report.txt", header = TRUE,check.names = FALSE, stringsAsFactors = FALSE) # NOTE the readin options
meta_table <- rio::import("./extdata/metainfo.txt", header = TRUE, check.names = FALSE, stringsAsFactors = FALSE) # with meta file

rmarkdown::render("./rmd/MQ_report_peptides.Rmd",output_format = "html_document", params = list(data_table =  data_table, meta_table = meta_table), output_file="MQ_report_peptides.html",output_dir = getwd())

remotely

without meta

url_api <- "http://206.12.91.148/ocpu/library/rmdocpu/R/render_peptides_file"

# get the root url
url_api_split <- strsplit(url_api, "/")[[1]]
url_server<- paste0(url_api_split[1],"//", url_api_split[3],"/")

# upload file and do the rendering
# variable r is the returning information from the curl function
r <- httr::POST(url_api, body = list(file = httr::upload_file("peptides.txt")))
#r <- httr::POST(url_api, body = list(file = httr::upload_file("peptides3.txt"),meta = httr::upload_file("peptides3_meta.txt")), httr::timeout(200000))
r$status_code

# get all the paths of all files from the opencpu end, and locate the one, which is the report
# this step needs to be done in the script enviroment

paths <- strsplit(rawToChar(r$content), "\n")[[1]]
path_target <- paths[grep("output.html",paths)]
path_target
# save/download the report file to local storage
# the file  "maxquant_result_summary.html" now is the report
curl::curl_download(paste0(url_server, path_target), "report_peptides_summary.html")

with meta

url_api <- "http://206.12.91.148/ocpu/library/rmdocpu/R/render_peptides_file"

# get the root url
url_api_split <- strsplit(url_api, "/")[[1]]
url_server<- paste0(url_api_split[1],"//", url_api_split[3],"/")

# upload file and do the rendering
# in this case, the proteinGroups.txt is in the working dir. it can be anywhere with the path
# variable r is the returning information from the curl function
#r <- httr::POST(url_api, body = list(file = httr::upload_file("peptides.txt")))
r <- httr::POST(url_api, body = list(file = httr::upload_file("peptides.txt"),meta = httr::upload_file("metainfo.txt")), httr::timeout(200000))
r$status_code

# get all the paths of all files from the opencpu end, and locate the one, which is the report
# this step needs to be done in the script enviroment

paths <- strsplit(rawToChar(r$content), "\n")[[1]]
path_target <- paths[grep("output.html",paths)]
path_target
# save/download the report file to local storage
# the file  "maxquant_result_summary.html" now is the report
curl::curl_download(paste0(url_server, path_target), "report_peptides_summary.html")

BuiltIn.taxa.refine.csv ----

remotely

without meta

url_api <- "http://206.12.91.148/ocpu/library/rmdocpu/R/render_taxon_file"

# get the root url
url_api_split <- strsplit(url_api, "/")[[1]]
url_server<- paste0(url_api_split[1],"//", url_api_split[3],"/")

# upload file and do the rendering
# variable r is the returning information from the curl function
r <- httr::POST(url_api, body = list(file = httr::upload_file("BuiltIn.taxa.refine.csv")))

# get all the paths of all files from the opencpu end, and locate the one, which is the report
# this step needs to be done in the script enviroment

paths <- strsplit(rawToChar(r$content), "\n")[[1]]
path_target <- paths[grep("output.html",paths)]

# save/download the report file to local storage
# the file  "maxquant_result_summary.html" now is the report
curl::curl_download(paste0(url_server, path_target), "report_taxonomy_summary.html")

with meta

url_api <- "http://206.12.91.148/ocpu/library/rmdocpu/R/render_taxon_file"

# get the root url
url_api_split <- strsplit(url_api, "/")[[1]]
url_server<- paste0(url_api_split[1],"//", url_api_split[3],"/")

# upload file and do the rendering
# variable r is the returning information from the curl function

r <- httr::POST(url_api, body = list(file = httr::upload_file("BuiltIn.taxa.refine.csv"),meta = httr::upload_file("metainfo.txt")), httr::timeout(200000))

# get all the paths of all files from the opencpu end, and locate the one, which is the report
# this step needs to be done in the script enviroment

paths <- strsplit(rawToChar(r$content), "\n")[[1]]
path_target <- paths[grep("output.html",paths)]

# save/download the report file to local storage
# the file  "maxquant_result_summary.html" now is the report
curl::curl_download(paste0(url_server, path_target), "report_taxonomy_summary.html")

function.csv ----

remote ----

without meta

url_api <- "http://206.12.91.148/ocpu/library/rmdocpu/R/render_function_file"

# get the root url
url_api_split <- strsplit(url_api, "/")[[1]]
url_server<- paste0(url_api_split[1],"//", url_api_split[3],"/")

# upload file and do the rendering
# variable r is the returning information from the curl function
r <- httr::POST(url_api, body = list(file = httr::upload_file("extdata/functions.csv")))

r$status_code

# get all the paths of all files from the opencpu end, and locate the one, which is the report
# this step needs to be done in the script enviroment

paths <- strsplit(rawToChar(r$content), "\n")[[1]]
path_target <- paths[grep("output.html",paths)]

# save/download the report file to local storage
# the file  "maxquant_result_summary.html" now is the report
curl::curl_download(paste0(url_server, path_target), "report_function_summary.html")

with meta

url_api <- "http://206.12.91.148/ocpu/library/rmdocpu/R/render_function_file"

# get the root url
url_api_split <- strsplit(url_api, "/")[[1]]
url_server<- paste0(url_api_split[1],"//", url_api_split[3],"/")

# upload file and do the rendering
# variable r is the returning information from the curl function
#r <- httr::POST(url_api, body = list(file = httr::upload_file("functions.csv")))
r <- httr::POST(url_api, body = list(file = httr::upload_file("functions.csv"), meta = httr::upload_file("metainfo.txt")))

r$status_code

# get all the paths of all files from the opencpu end, and locate the one, which is the report
# this step needs to be done in the script enviroment

paths <- strsplit(rawToChar(r$content), "\n")[[1]]
path_target <- paths[grep("output.html",paths)]

# save/download the report file to local storage
# the file  "maxquant_result_summary.html" now is the report
curl::curl_download(paste0(url_server, path_target), "report_function_summary.html")

locally

without meta

data_table <- read.delim("extdata/functions.csv", header = TRUE,sep= ",",check.names = TRUE, stringsAsFactors = FALSE)
#data_table <- read.delim("extdata/functions_20211015.csv", header = TRUE,sep= ",",check.names = TRUE, stringsAsFactors = FALSE)
# knit to all supported format (here we support pdf and html)

data_table <- read.delim("extdata/functions_20211018.tsv", header = TRUE,sep= "\t",quote="", check.names = TRUE, stringsAsFactors = FALSE,)



data_table <- rio::import("extdata/functions_20211018.tsv",check.names = TRUE)



rmarkdown::render("rmd/ML_report_function.Rmd",output_format = "html_document", params = list(input_datatable =  data_table))

with meta

### with meta

data_table <- read.delim("functions.csv", header = TRUE,check.names = FALSE, stringsAsFactors = FALSE)
meta_table <- read.delim("metainfo.txt", header = TRUE,check.names = FALSE, stringsAsFactors = FALSE) 

rmarkdown::render("ML_report_function.Rmd",output_format = "html_document", params = list(input_datatable =  data_table, meta_table = meta_table ))

pepFunk

locally

data_table  <- data_table <- read.delim("./extdata/peptides.txt", row.names = 1) # NOTE the readin options

meta_table <- rio::import("./extdata/metainfo_pepfunk.txt", header = TRUE, check.names = FALSE, stringsAsFactors = FALSE) # with meta file

#meta_table <- read.delim("./extdata/metadata.csv", header=F, sep=",")


rmarkdown::render("./rmd/ML_report_pepFunk.Rmd",output_format = "html_document", output_dir = getwd(), params = list(data_table =  data_table, meta_table = meta_table ))



rmarkdown::render("./rmd/ML_report_pepFunk.Rmd",output_format = "html_document", output_dir = getwd(), envir = new.env(), params = list(data_table =  data_table, meta_table = meta_table ))



# for debugging

#params$data_table <- data_table
#params$meta_table <- meta_table

ningzhibin/rmdocpu documentation built on Feb. 3, 2022, 9:30 p.m.