re-allow samples = NULL in combineBCR

R/combineContigs.R

@@ -1,5 +1,9 @@
 # Adding Global Variables
 # data('v_gene','j_gene', 'c_gene', 'd_gene')
+# note that currently the Rcpp internals have hardcoded column names so
+# if some breaking change here is made, the Rcpp code will need to be updated,
+# or functions need to be adjusted to intake expected column names that
+# uses these variables
 utils::globalVariables(c("v_gene", "j_gene", "c_gene", "d_gene", "chain"))
 
 heavy_lines <- c("IGH", "cdr3_aa1", "cdr3_nt1", "vgene1")
@@ -21,42 +25,42 @@ utils::globalVariables(c(
 #' @title Combining the list of T cell receptor contigs into clones
 #'
 #' @description This function consolidates a list of TCR sequencing results to
-#' the level of  the individual cell barcodes. Using the **samples** and 
-#' **ID** parameters, the function will add the strings as prefixes to 
-#' prevent issues with repeated  barcodes. The resulting new barcodes will 
-#' need to match the Seurat or SCE object in order to use, 
-#' [combineExpression()]. Several levels of filtering exist - 
-#' *removeNA*, *removeMulti*, or *filterMulti* are parameters 
-#' that control how the function deals with barcodes with multiple chains 
+#' the level of  the individual cell barcodes. Using the **samples** and
+#' **ID** parameters, the function will add the strings as prefixes to
+#' prevent issues with repeated  barcodes. The resulting new barcodes will
+#' need to match the Seurat or SCE object in order to use,
+#' [combineExpression()]. Several levels of filtering exist -
+#' *removeNA*, *removeMulti*, or *filterMulti* are parameters
+#' that control how the function deals with barcodes with multiple chains
 #' recovered.
-#' 
+#'
 #' @examples
-#' combined <- combineTCR(contig_list, 
-#'                         samples = c("P17B", "P17L", "P18B", "P18L", 
+#' combined <- combineTCR(contig_list,
+#'                         samples = c("P17B", "P17L", "P18B", "P18L",
 #'                                     "P19B","P19L", "P20B", "P20L"))
-#' 
-#' @param input.data List of filtered contig annotations or 
+#'
+#' @param input.data List of filtered contig annotations or
 #' outputs from [loadContigs()].
 #' @param samples The labels of samples (recommended).
 #' @param ID The additional sample labeling (optional).
 #' @param removeNA This will remove any chain without values.
 #' @param removeMulti This will remove barcodes with greater than 2 chains.
-#' @param filterMulti This option will allow for the selection of the 2 
-#' corresponding chains with the highest expression for a single barcode. 
-#' @param filterNonproductive This option will allow for the removal of 
+#' @param filterMulti This option will allow for the selection of the 2
+#' corresponding chains with the highest expression for a single barcode.
+#' @param filterNonproductive This option will allow for the removal of
 #' nonproductive chains if the variable exists in the contig data. Default
 #' is set to TRUE to remove nonproductive contigs.
-#' 
+#'
 #' @importFrom assertthat assert_that is.flag
 #' @export
 #' @concept Loading_and_Processing_Contigs
 #' @return List of clones for individual cell barcodes
-#' 
-combineTCR <- function(input.data, 
-                       samples = NULL, 
-                       ID = NULL, 
-                       removeNA = FALSE, 
-                       removeMulti = FALSE, 
+#'
+combineTCR <- function(input.data,
+                       samples = NULL,
+                       ID = NULL,
+                       removeNA = FALSE,
+                       removeMulti = FALSE,
                        filterMulti = FALSE,
                        filterNonproductive = TRUE) {
 
@@ -66,7 +70,7 @@ combineTCR <- function(input.data,
     assert_that(is.flag(removeNA))
     assert_that(is.flag(removeMulti))
     assert_that(is.flag(filterMulti))
-    
+
     input.data <- .checkList(input.data)
     input.data <- .checkContigs(input.data)
     out <- NULL
@@ -80,8 +84,8 @@ combineTCR <- function(input.data,
         }
         input.data[[i]]$sample <- samples[i]
         input.data[[i]]$ID <- ID[i]
-        if (filterMulti) { 
-          input.data[[i]] <- .filteringMulti(input.data[[i]]) 
+        if (filterMulti) {
+          input.data[[i]] <- .filteringMulti(input.data[[i]])
         }
     }
     #Prevents error caused by list containing elements with 0 rows
@@ -104,10 +108,10 @@ combineTCR <- function(input.data,
         data2 <- .makeGenes(cellType = "T", out[[i]])
         Con.df <- .constructConDfAndParseTCR(data2)
         Con.df <- .assignCT(cellType = "T", Con.df)
-        Con.df[Con.df == "NA_NA" | Con.df == "NA;NA_NA;NA"] <- NA 
-        data3 <- merge(data2[,-which(names(data2) %in% c("TCR1","TCR2"))], 
+        Con.df[Con.df == "NA_NA" | Con.df == "NA;NA_NA;NA"] <- NA
+        data3 <- merge(data2[,-which(names(data2) %in% c("TCR1","TCR2"))],
             Con.df, by = "barcode")
-        
+
         columns_to_include <- c("barcode")
         # Conditionally add columns based on user input
         if (!is.null(samples)) {
@@ -116,17 +120,17 @@ combineTCR <- function(input.data,
         if (!is.null(ID)) {
           columns_to_include <- c(columns_to_include, "ID")
         }
-        
+
         # Add TCR and CT lines which are presumably always needed
         columns_to_include <- c(columns_to_include, tcr1_lines, tcr2_lines, CT_lines)
-        
+
         # Subset the data frame based on the dynamically built list of columns
         data3 <- data3[, columns_to_include]
-        
-        final[[i]] <- data3 
+
+        final[[i]] <- data3
     }
     name_vector <- character(length(samples))
-    for (i in seq_along(samples)) { 
+    for (i in seq_along(samples)) {
         if (!is.null(samples) && !is.null(ID)) {
             curr <- paste(samples[i], "_", ID[i], sep="")
         } else if (!is.null(samples) & is.null(ID)) {
@@ -155,61 +159,66 @@ combineTCR <- function(input.data,
 
 #' Combining the list of B cell receptor contigs into clones
 #'
-#' This function consolidates a list of BCR sequencing results to the level 
-#' of the individual cell barcodes. Using the samples and ID parameters, 
-#' the function will add the strings as prefixes to prevent issues with 
-#' repeated barcodes. The resulting new barcodes will need to match the 
-#' Seurat or SCE object in order to use, [combineExpression()]. 
-#' Unlike [combineTCR()], combineBCR produces a column 
-#' **CTstrict** of an index of nucleotide sequence and the 
-#' corresponding V gene. This index automatically calculates the 
+#' This function consolidates a list of BCR sequencing results to the level
+#' of the individual cell barcodes. Using the samples and ID parameters,
+#' the function will add the strings as prefixes to prevent issues with
+#' repeated barcodes. The resulting new barcodes will need to match the
+#' Seurat or SCE object in order to use, [combineExpression()].
+#' Unlike [combineTCR()], combineBCR produces a column
+#' **CTstrict** of an index of nucleotide sequence and the
+#' corresponding V gene. This index automatically calculates the
 #' Levenshtein distance between sequences with the same V gene and will
-#' index sequences using a normalized Levenshtein distance with the same 
-#' ID. After which, clone clusters are called using the 
-#' [igraph::components()] function. Clones that are clustered 
-#' across multiple sequences will then be labeled with "Cluster" in the 
+#' index sequences using a normalized Levenshtein distance with the same
+#' ID. After which, clone clusters are called using the
+#' [igraph::components()] function. Clones that are clustered
+#' across multiple sequences will then be labeled with "Cluster" in the
 #' CTstrict header.
 #'
 #' @examples
 #' #Data derived from the 10x Genomics intratumoral NSCLC B cells
 #' BCR <- read.csv("https://www.borch.dev/uploads/contigs/b_contigs.csv")
-#' combined <- combineBCR(BCR, 
-#'                        samples = "Patient1", 
+#' combined <- combineBCR(BCR,
+#'                        samples = "Patient1",
 #'                        threshold = 0.85)
-#' 
-#' @param input.data List of filtered contig annotations or outputs from 
+#'
+#' @param input.data List of filtered contig annotations or outputs from
 #' [loadContigs()].
 #' @param samples The labels of samples (required).
 #' @param ID The additional sample labeling (optional).
-#' @param call.related.clones Use the nucleotide sequence and V gene 
-#' to call related clones. Default is set to TRUE. FALSE will return 
+#' @param call.related.clones Use the nucleotide sequence and V gene
+#' to call related clones. Default is set to TRUE. FALSE will return
 #' a CTstrict or strict clone as V gene + amino acid sequence.
-#' @param threshold The normalized edit distance to consider. The higher 
+#' @param threshold The normalized edit distance to consider. The higher
 #' the number the more similarity of sequence will be used for clustering.
 #' @param removeNA This will remove any chain without values.
 #' @param removeMulti This will remove barcodes with greater than 2 chains.
-#' @param filterMulti This option will allow for the selection of the 
+#' @param filterMulti This option will allow for the selection of the
 #' highest-expressing light and heavy chains, if not calling related clones.
-#' @param filterNonproductive This option will allow for the removal of 
+#' @param filterNonproductive This option will allow for the removal of
 #' nonproductive chains if the variable exists in the contig data. Default
 #' is set to TRUE to remove nonproductive contigs.
 #'
 #' @export
 #' @concept Loading_and_Processing_Contigs
 #' @return List of clones for individual cell barcodes
 combineBCR <- function(input.data,
-                       samples,
+                       samples = NULL,
                        ID = NULL,
                        call.related.clones = TRUE,
                        threshold = 0.85,
-                       removeNA = FALSE, 
+                       removeNA = FALSE,
                        removeMulti = FALSE,
                        filterMulti = TRUE,
                        filterNonproductive = TRUE) {
 
+    if (is.null(samples)) {
+        stop("combineBCR() requires the samples parameter for the calculation of edit distance.")
+    }
+
     assert_that(
-        isListOfNonEmptyDataFrames(input.data) || isNonEmptyDataFrame(input.data),
-        is.character(samples),
+        isListOfNonEmptyDataFrames(input.data) ||
+            isNonEmptyDataFrame(input.data),
+        is.character(samples) || is.null(samples),
         is.flag(call.related.clones),
         is.numeric(threshold),
         is.flag(removeNA),
@@ -220,7 +229,8 @@ combineBCR <- function(input.data,
     final <- input.data %>%
         .checkList() %>%
         .checkContigs() %>%
-        lapply(function(x) {
+        unname() %>%
+        purrr::imap(function(x, i) {
             x <- subset(x, chain %in% c("IGH", "IGK", "IGL"))
             if (!is.null(ID)) x$ID <- ID[i]
             if (filterNonproductive && "productive" %in% colnames(x)) {
@@ -230,15 +240,21 @@ combineBCR <- function(input.data,
                 # Keep IGH / IGK / IGL info in save_chain
                 x$save_chain <- x$chain
                 # Collapse IGK and IGL chains
-                x$chain <- ifelse(x$chain=="IGH","IGH","IGLC")
+                x$chain <- ifelse(x$chain == "IGH", "IGH", "IGLC")
                 x <- .filteringMulti(x)
                 # Get back IGK / IGL distinction
                 x$chain <- x$save_chain
                 x$save_chain <- NULL
             }
             x
         }) %>%
-        .modifyBarcodes(samples, ID) %>%
+        (function(x) {
+            if (!is.null(samples)) {
+                .modifyBarcodes(x, samples, ID)
+            } else { # https://github.com/ncborcherding/scRepertoire/pull/450
+                x
+            }
+        }) %>%
         lapply(function(x) {
             data2 <- data.frame(x)
             data2 <- .makeGenes(cellType = "B", data2)
@@ -270,26 +286,23 @@ combineBCR <- function(input.data,
       }
         final[[i]]$sample <- samples[i]
         final[[i]]$ID <- ID[i]
-        final[[i]][final[[i]] == "NA_NA" | final[[i]] == "NA;NA_NA;NA"] <- NA 
+        final[[i]][final[[i]] == "NA_NA" | final[[i]] == "NA;NA_NA;NA"] <- NA
         if (!is.null(sample) & !is.null(ID)) {
-          final[[i]]<- final[[i]][, c("barcode", "sample", "ID", 
+          final[[i]]<- final[[i]][, c("barcode", "sample", "ID",
               heavy_lines[c(1,2,3)], light_lines[c(1,2,3)], CT_lines)]
         }
         else if (!is.null(sample) & is.null(ID)) {
-          final[[i]]<- final[[i]][, c("barcode", "sample", 
+          final[[i]]<- final[[i]][, c("barcode", "sample",
                     heavy_lines[c(1,2,3)], light_lines[c(1,2,3)], CT_lines)]
         }
     }
-    names <- NULL
-    for (i in seq_along(samples)) {
-        if (!is.null(samples) && !is.null(ID)) {
-            c <- paste(samples[i], "_", ID[i], sep="")
-        } else if (!is.null(samples) && is.null(ID)) {
-            c <- paste(samples[i], sep = "")
-        }
-        names <- c(names, c)
+
+    names(final) <- if (!is.null(samples)) {
+        if (is.null(ID)) samples else paste0(samples, "_", ID)
+    } else {
+        paste0("S", seq_along(final))
     }
-    names(final) <- names
+
     for (i in seq_along(final)) {
         final[[i]] <- final[[i]][!duplicated(final[[i]]$barcode),]
         final[[i]] <- final[[i]][rowSums(is.na(final[[i]])) < 10, ]