Merge branch 'release-1.2.0'

.Rbuildignore

@@ -21,4 +21,5 @@ runtime.txt
 .github
 .git
 Aux
-binder
+binder
+tests

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: DIscBIO
-Date: 2020-11-13
+Date: 2021-04-28
 Title: A User-Friendly Pipeline for Biomarker Discovery in Single-Cell Transcriptomics
-Version: 1.1.0
+Version: 1.2.0
 Authors@R:
 	c(
 		person(
@@ -32,14 +32,15 @@ Description: An open, multi-algorithmic pipeline for easy, fast and efficient
 	analysis of cellular sub-populations and the molecular signatures that
 	characterize them. The pipeline consists of four successive steps: data
 	pre-processing, cellular clustering with pseudo-temporal ordering, defining
-	differential expressed genes and biomarker identification. This package
-	implements extensions of the work published by Ghannoum et. al. (2019)
+	differential expressed genes and biomarker identification. More details on
+	Ghannoum et. al. (2021) <doi:10.3390/ijms22031399>. This package implements
+	extensions of the work published by Ghannoum et. al. (2019)
 	<doi:10.1101/700989>.
 License: MIT + file LICENSE
 Encoding: UTF-8
 Imports: methods, TSCAN, boot, httr, mclust, statmod, igraph,
 	RWeka, philentropy, NetIndices, png, grDevices,
-	readr, RColorBrewer, ggplot2, rpart, fpc, cluster, rpart.plot,
+	RColorBrewer, ggplot2, rpart, fpc, cluster, rpart.plot,
 	tsne, AnnotationDbi, org.Hs.eg.db, graphics, stats, utils, impute
 Depends:
 	R (>= 4.0), SingleCellExperiment

NAMESPACE

@@ -113,7 +113,6 @@ importFrom(methods,new)
 importFrom(methods,validObject)
 importFrom(philentropy,distance)
 importFrom(png,readPNG)
-importFrom(readr,read_tsv)
 importFrom(rpart,rpart)
 importFrom(rpart.plot,rpart.plot)
 importFrom(statmod,glmgam.fit)

NEWS.md

@@ -1,5 +1,20 @@
 This file documents the changes between stable releases of DIscBIO.
 
+# Version 1.2.0
+
+- Added support for retrying to retrieve URLs (issue #29)
+- Removed duplicated (legacy) function
+- Improved detection of duplicated data on `NetAnalysis()`
+- Improved I/O behavior
+- Improved validation
+- Bug fixes
+
+# Version 1.1.0
+
+- Added support for more gene types
+- Improved I/O behavior
+- Updated Binder notebooks
+
 # Version 1.0.1
 
 - Bug fixes

R/DIscBIO-generic-DEGanalysis2clust.R

@@ -91,17 +91,19 @@ setMethod(
         x <- L
         data = list(x = x, y = y, geneid = gname)
         if (quiet) {
-            invisible(capture.output({
-                samr.obj <- sammy(
-                    data,
-                    resp.type = "Two class unpaired",
-                    assay.type = "seq",
-                    testStatistic = "wilcoxon",
-                    random.seed = 15,
-                    ...
-                )
-                delta.table <- samr.compute.delta.table(samr.obj)
-            }))
+            suppressMessages(
+                invisible(capture.output({
+                    samr.obj <- sammy(
+                        data,
+                        resp.type = "Two class unpaired",
+                        assay.type = "seq",
+                        testStatistic = "wilcoxon",
+                        random.seed = 15,
+                        ...
+                    )
+                    delta.table <- samr.compute.delta.table(samr.obj)
+                }))
+            )
         } else {
             samr.obj <- sammy(
                 data,

R/DIscBIO-generic-PlotmclustMB.R

@@ -51,7 +51,7 @@ setMethod(
                 )
                 if (showcluster) {
                     g <- g + geom_point(
-                        aes_string(color = color_by), na.rm = TRUE
+                        aes_string(color = lib_info_with_pseudo$State), na.rm = TRUE
                     )
                     g <- g + scale_colour_manual(
                         values = c(
@@ -77,7 +77,7 @@ setMethod(
                     )
                 if (showcluster) {
                     g <- g + geom_point(
-                        aes_string(color = color_by), na.rm = TRUE, size = 3
+                        aes_string(color = lib_info_with_pseudo$State), na.rm = TRUE, size = 3
                     )
                     g <- g + scale_colour_manual(
                         values = c(
@@ -224,4 +224,4 @@ setMethod(
         }
         Plotmclust(total)
     }
-)
+)

R/NetAnalysis.R

@@ -23,6 +23,12 @@ NetAnalysis <- function(
     names <- rownames(betweenness.table)
     rownames(betweenness.table) <- NULL
     degree <- degree(gg)
+
+    Duplicated<-data[duplicated(data),]
+    if(length(Duplicated[,1])>0){
+        degree=degree/2
+    }
+
     degree.table <- data.frame(degree)
     names <- rownames(degree.table)
     rownames(degree.table) <- NULL

R/Networking.R

@@ -1,5 +1,6 @@
 #' @title Plotting the network.
-#' @description This function uses STRING-api to plot the network.
+#' @description This function uses STRING API to plot the network.
+#' @references https://string-db.org/api/
 #' @export
 #' @param data A gene list.
 #' @param FileName A string vector showing the name to be used to save the
@@ -8,77 +9,55 @@
 #' @param species The taxonomy name/id. Default is "9606" for Homo sapiens.
 #' @param plot_width Plot width
 #' @param plot_height Plot height
-#' @importFrom httr GET status_code
+#' @param retries maximum number of attempts to connect to the STRING api.
 #' @importFrom utils download.file
 #' @importFrom png readPNG
 #' @importFrom graphics plot rasterImage
 #' @return A plot of the network
 Networking <- function(
-    data,
-    FileName    = NULL,
-    species     = "9606",
-    plot_width  = 25,
-    plot_height = 15
-    )
-    {
-        if (length(data) > 600) {
-            print("Your gene list is too big")
-        } else{
-            string_api_url <- "https://string-db.org/api/"
-            output_format <- "highres_image"
-            method <- "network"
-            your_identifiers <- ""
-            optional_parameters <- ""
+	data,
+	FileName    = NULL,
+	species     = "9606",
+	plot_width  = 25,
+	plot_height = 15,
+	retries     = 3
+) {
 
-            # Construct API request
-            genes <- data
-            repos <-
-                GET(
-                    url = paste0(
-                        string_api_url,
-                        output_format,
-                        '/',
-                        method,
-                        '?identifiers=',
-                        paste(as.character(data), collapse = "%0d"),
-                        "&",
-                        "species=",
-                        species
-                    )
-                )
-            message(
-                "Examine response components =",
-                status_code(repos),
-                "\t",
-                "(200 means successful)"
-            )
-            y <- repos$request$url
-            if (!is.null(FileName)) {
-                FileName <- paste0("network", FileName, ".png")
-            } else {
-                FileName <- tempfile()
-            }
-            download.file(y, FileName, mode = 'wb')
-            message(
-                "\n",
-                "You can see the network with high resolution ",
-                "by clicking on the following link:",
-                "\n",
-                paste0(y)
-            )
-            Network <- readPNG(FileName, native=TRUE)
-            set_plot_dimensions <-
-                function(width_choice, height_choice) {
-                    opar <- options(
-                        repr.plot.width = width_choice,
-                        repr.plot.height = height_choice
-                    )
-                    on.exit(options(opar))
-                }
-            set_plot_dimensions(plot_width, plot_height)
+	# ======================================================== #
+	# Validation                                               #
+	# ======================================================== #
+	if (length(data) > 600) stop("Your gene list is too big")
 
-            plot(0:1, 0:1, type = "n", ann = FALSE, axes = FALSE)
-            rasterImage(Network, 0, 0, 1, 1)
-            set_plot_dimensions(8, 8) # resets to default values
-        }
-    }
+	# ======================================================== #
+	# Processing                                               #
+	# ======================================================== #
+	repos <- retrieveURL(data, species, "highres_image")
+	y <- repos$request$url
+	if (!is.null(FileName)) {
+		FileName <- paste0("network", FileName, ".png")
+	} else {
+		FileName <- tempfile()
+	}
+	download.file(y, FileName, mode = 'wb')
+	message(
+		"\n",
+		"You can see the network with high resolution ",
+		"by clicking on the following link:",
+		"\n",
+		paste0(y)
+	)
+	Network <- readPNG(FileName, native=TRUE)
+	set_plot_dimensions <- function(width_choice, height_choice) {
+		opar <- options(
+			repr.plot.width = width_choice,
+			repr.plot.height = height_choice
+		)
+		on.exit(options(opar))
+	}
+	set_plot_dimensions(plot_width, plot_height)
+
+	# Plotting ----------------------------------------------- #
+	plot(0:1, 0:1, type = "n", ann = FALSE, axes = FALSE)
+	rasterImage(Network, 0, 0, 1, 1)
+	set_plot_dimensions(8, 8) # resets to default values
+}

R/PPI.R

@@ -1,50 +1,23 @@
 #' @title Defining protein-protein interactions (PPI) over a list of genes,
 #' @description This function uses STRING-api. The outcome of STRING analysis
-#'   will be stored in tab separated values (TSV) files.
+#'   will be stored in comma-separated values files.
 #' @export
 #' @param data A gene list.
 #' @param FileName A string vector showing the name to be used to save the
 #'   resulted table. If null, no file will be exported
 #' @param species The taxonomy name/id. Default is "9606" for Homo sapiens.
 #' @importFrom httr content
-#' @importFrom readr read_tsv
 #' @importFrom utils read.table write.table
-#' @return Either a TSV file stored in the user's file system and its
+#' @return Either CSV files stored in the user's file system and its
 #' corresponding `data.frame` object in R or and R object containing that
 #' information.
 PPI <- function(data, FileName = NULL, species = "9606") {
-    # Save base enpoint as variable
-    string_api_url <- "https://string-db.org/api/"
-    output_format <- "tsv" #"json", "tsv-no-header", "tsv", "xml"
-    method <- "network"
-    your_identifiers <- ""
-    optional_parameters <- ""
-    # Construct API request
-    genes <- data
-    repos <- GET(
-        url = paste0(
-            string_api_url,
-            output_format,
-            '/',
-            method,
-            '?identifiers=',
-            paste(as.character(data), collapse = "%0d"),
-            "&",
-            "species=",
-            species
-        )
-    )
-    message(
-        "Examine response components = ", status_code(repos), "\t",
-        "(200 means successful)", "\n"
-    )
-    # Process API request content
-    repo_content <- content(repos)
-    #results <- read_tsv(repo_content)
-    write.table(repo_content, file = "data.csv", sep = ",")
-    results <- read.table(file = "data.csv", sep = ",")
-    if (!is.null(FileName)) {
-        write.csv(results, file = paste0("PPI-", FileName, ".csv"))
-    }
-    return(results)
+	repos <- retrieveURL(data, species, "tsv")
+	# Process API request content
+	repo_content <- content(repos)
+	results <- data.frame(repo_content[, , 1])
+	if (!is.null(FileName)) {
+		write.csv(results, file = paste0("PPI-", FileName, ".csv"))
+	}
+	return(results)
 }

R/internal-functions.R

@@ -365,3 +365,57 @@ prepExampleDataset <- function(dataset, save=TRUE) {
 		message("Not saving dataset because (save == FALSE)")
 	}
 }
+
+#' @title Retries a URL
+#' @description Retries a URL
+#' @param data A gene list
+#' @param species The taxonomy name/id. Default is "9606" for Homo sapiens
+#' @param outputFormat format of the output. Can be "highres_image", "tsv",
+#' "json", "tsv-no-header", "xml"
+#' @param maxRetries maximum number of attempts to connect to the STRING api.
+#' @param successCode Status code number that represents success
+#' @return either the output of httr::GET or an error message
+#' @importFrom httr GET status_code
+#' @author Waldir Leoncio
+retrieveURL <- function(
+	data, species, outputFormat, maxRetries=3, successCode=200
+) {
+	# ======================================================== #
+	# Setting up retrieval                                     #
+	# ======================================================== #
+	string_api_url <- "https://string-db.org/api/"
+	method <- "network"
+	url <- paste0(
+		string_api_url, outputFormat, '/', method, '?identifiers=',
+		paste(as.character(data), collapse = "%0d"), "&species=",
+		species
+	)
+
+	# ======================================================== #
+	# Retrieving URL                                           #
+	# ======================================================== #
+	message("Retrieving URL. Please wait...")
+	repos <- GET(url)
+	failedGET <- status_code(repos) != successCode
+	r <- 1
+	while (failedGET & (r <= maxRetries)) {
+		message("Failed retrieval. Retry ", r, " out of ", maxRetries, ".")
+		repos <- GET(url)
+		failedGET <- status_code(repos) != successCode
+		r <- r + 1
+	}
+
+	# ======================================================== #
+	# Final output                                             #
+	# ======================================================== #
+	if (failedGET) {
+		stop(
+			"Unable to retrieve URL. Please check the parameters ",
+			"passed to the Networking() function, increase the ",
+			"'maxRetries' parameter or try again later."
+		)
+	} else {
+		message("Successful retrieval.")
+		return(repos)
+	}
+}