Merge pull request #45 from CostaLab/devel

Updated namespace
CostaLab · Oct 15, 2024 · e880916 · e880916
2 parents d514d9a + a32ad59
commit e880916
Show file tree

Hide file tree

Showing 8 changed files with 50 additions and 17 deletions.
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -1,7 +1,7 @@
 Package: sigurd
 Type: Package
 Title: Single cell Genotyping Using RNA Data
-Version: 0.3.4
+Version: 0.3.6
 Authors@R: c(
   person(given = "Martin",
          family = "Grasshoff",

diff --git a/NAMESPACE b/NAMESPACE
@@ -8,8 +8,10 @@ export(CalculateConsensus)
 export(CalculateCorrelationPValue)
 export(CalculateCoverage)
 export(CalculateFisherTestPValue)
+export(CalculateForwardReads)
 export(CalculateQuality)
 export(CalculateRefReads)
+export(CalculateReverseReads)
 export(CalculateStrandCorrelation)
 export(CallSupport)
 export(ClonalDefinition)
@@ -21,6 +23,7 @@ export(GetCellInfoPerVariant)
 export(GetVariantInfo)
 export(HeatmapVoi)
 export(LoadingMAEGATK_typewise)
+export(LoadingMGATK_typewise)
 export(LoadingRawMatrix_typewise)
 export(LoadingVCF_typewise)
 export(LoadingVarTrix_typewise)
@@ -38,6 +41,7 @@ export(VariantQuantileThresholding_Combined)
 export(VariantSelection_Group)
 export(VariantSelection_Quantile)
 export(VariantSelection_TopCells)
+export(VariantSelection_VMR)
 export(VariantWiseCorrelation)
 export(VariantWiseFisherTest)
 export(char_to_numeric)
@@ -70,6 +74,7 @@ importFrom(Matrix,rowSums)
 importFrom(Matrix,sparseMatrix)
 importFrom(Matrix,summary)
 importFrom(Matrix,t)
+importFrom(MatrixGenerics,rowVars)
 importFrom(S4Vectors,DataFrame)
 importFrom(S4Vectors,merge)
 importFrom(S4Vectors,metadata)

diff --git a/R/RowWiseSplit.R b/R/RowWiseSplit.R
@@ -3,15 +3,19 @@
 #'Performing the correlation or Fisher test association for a SummarizedExperiment object requires extreme amounts of memory.
 #'To reduce the amount of memory necessary, we instead get the individual rows from the consensus assay.
 #'We can then remove the NoCalls (no reads) from the individual vectors, further reducing the amount of memory needed.
+#'
+#'When the NoCalls are removed, the repective cells are also removed if a different assay is selected.
 #'@importFrom parallel mclapply
 #'@importFrom SummarizedExperiment assays
 #'@param se SummarizedExperiment object.
 #'@param n_cores Number of cores to use.
 #'@param remove_nocalls Do you want to remove NoCall cells?
+#'@param assay_to_split Which assay to you want to split?
 #'@export
-RowWiseSplit <- function(se, n_cores = 1, minimum_reads, remove_nocalls = TRUE){
-  consensus <- SummarizedExperiment::assays(se)$consensus
-  consensus_list <- parallel::mclapply(rownames(se), SeparatingMatrixToList, total_matrix = consensus, remove_nocalls = remove_nocalls, mc.cores = n_cores)
+RowWiseSplit <- function(se, n_cores = 1, assay_to_split = "consensus", remove_nocalls = TRUE){
+  selected_assay <- SummarizedExperiment::assays(se)[[assay_to_split]]
+  consensus <- SummarizedExperiment::assays(se)[["consensus"]]
+  consensus_list <- parallel::mclapply(rownames(se), SeparatingMatrixToList, total_matrix = selected_assay, assay_to_split = assay_to_split, consensus = consensus, remove_nocalls = remove_nocalls, mc.cores = n_cores)
   names(consensus_list) <- rownames(se)
   return(consensus_list)
 }
diff --git a/R/SeparatingMatrixToList.R b/R/SeparatingMatrixToList.R
@@ -7,20 +7,27 @@
 #'@importFrom stats na.omit
 #'@param row_use The row the separate.
 #'@param total_matrix The matrix to be split.
+#'@param assay_to_split Which assay are we splitting?
 #'@param remove_nocalls Do you want to remove NoCall cells?
 #'@export
-SeparatingMatrixToList <- function(row_use, total_matrix, remove_nocalls = TRUE){
+SeparatingMatrixToList <- function(row_use, total_matrix, consensus, assay_to_split = "consensus", remove_nocalls = TRUE){
   selected_row <- total_matrix[row_use,]
   selected_row <- stats::na.omit(selected_row)
 
   if(remove_nocalls == TRUE){
     # We remove the NoCall cells.
-    selected_row <- selected_row[selected_row != 0]
-    selected_row[selected_row == 1] <- 0
-    selected_row[selected_row >= 2] <- 1
+    selected_consensus <- consensus[row_use,]
+    selected_consensus <- stats::na.omit(selected_consensus)
+    selected_row <- selected_row[selected_consensus != 0]
+    if(assay_to_split == "consensus"){
+      selected_row[selected_row == 1] <- 0
+      selected_row[selected_row >= 2] <- 1
+    }
   } else if(remove_nocalls == FALSE){
-    selected_row[selected_row <= 1] <- 0
-    selected_row[selected_row >= 2] <- 1
+    if(assay_to_split == "consensus"){
+      selected_row[selected_row <= 1] <- 0
+      selected_row[selected_row >= 2] <- 1
+    }
   }
   return(selected_row)
 }
diff --git a/README.md b/README.md
@@ -33,7 +33,7 @@ The mutation data was obtained from the Sanger Institute Catalogue Of Somatic Mu
 
 ```
 
-# Current Features v0.3.4
+# Current Features v0.3.6
 
 - Loading data from VarTrix and MAEGATK.
 - Transforming the data to be compatible for joint analysis.

diff --git a/man/RowWiseSplit.Rd b/man/RowWiseSplit.Rd
diff --git a/man/SeparatingMatrixToList.Rd b/man/SeparatingMatrixToList.Rd
diff --git a/tests/testthat/test-SeparatingMatrixToList.R b/tests/testthat/test-SeparatingMatrixToList.R
@@ -5,10 +5,10 @@ test_that("Testing SeparatingMatrixToList.R", {
   test2 <- sigurd::SeparatingMatrixToList("chrM_2_A_G", input_matrix, remove_nocalls = FALSE)
   test3 <- sigurd::SeparatingMatrixToList("chrM_2_A_T", input_matrix, remove_nocalls = FALSE)
   test4 <- sigurd::SeparatingMatrixToList("chrM_4_C_G", input_matrix, remove_nocalls = FALSE)
-  test1_removed_nocalls <- sigurd::SeparatingMatrixToList("chrM_1_G_A", input_matrix, remove_nocalls = TRUE)
-  test2_removed_nocalls <- sigurd::SeparatingMatrixToList("chrM_2_A_G", input_matrix, remove_nocalls = TRUE)
-  test3_removed_nocalls <- sigurd::SeparatingMatrixToList("chrM_2_A_T", input_matrix, remove_nocalls = TRUE)
-  test4_removed_nocalls <- sigurd::SeparatingMatrixToList("chrM_4_C_G", input_matrix, remove_nocalls = TRUE)
+  test1_removed_nocalls <- sigurd::SeparatingMatrixToList("chrM_1_G_A", input_matrix, consensus = input_matrix, remove_nocalls = TRUE)
+  test2_removed_nocalls <- sigurd::SeparatingMatrixToList("chrM_2_A_G", input_matrix, consensus = input_matrix, remove_nocalls = TRUE)
+  test3_removed_nocalls <- sigurd::SeparatingMatrixToList("chrM_2_A_T", input_matrix, consensus = input_matrix, remove_nocalls = TRUE)
+  test4_removed_nocalls <- sigurd::SeparatingMatrixToList("chrM_4_C_G", input_matrix, consensus = input_matrix, remove_nocalls = TRUE)
   # We generate the expected outputs.
   expected_result1 <- c(Cell_1 = 0, Cell_2 = 1, Cell_3 = 1, Cell_4 = 0)
   expected_result2 <- c(Cell_1 = 0, Cell_2 = 1, Cell_3 = 1, Cell_4 = 0)