Merge branch 'main' into docs

bin/HTODemux.R

@@ -58,11 +58,7 @@ assignment <- hashtag[[paste0(args$assay,"_classification")]]
 assignment[[paste0(args$assay,"_classification")]][!assignment[[paste0(args$assay,"_classification")]] %in% c(donors, 'Negative')] <- "Doublet"
 
 
-print("------------------- Following Files are saved ----------------------------")
-print(paste0(args$assignmentOutHTOdemux, "_assignment_htodemux.csv"))
-print(paste0(args$assignmentOutHTOdemux, "_classification_htodemux.csv"))
-print(paste0(args$objectOutHTOdemux,".rds"))
-print("params.csv")
+
 write.csv(params, paste0(args$outputdir, "/params.csv"))
 write.csv(assignment, paste0(args$outputdir, "/", args$assignmentOutHTOdemux, "_assignment_htodemux.csv"))
 #write.csv(hashtag[[paste0(args$assay,"_classification")]], paste0(args$outputdir, "/", args$assignmentOutHTOdemux, "_assignment_htodemux.csv"))

bin/MultiSeq.R

@@ -58,11 +58,7 @@ hashtag[[args$assay]]
 print("-----------------------------------------------")
 colnames(x = hashtag[[]])
 
-#Save Results
-print("------------------- Following Files are saved ----------------------------")
-print(paste0(args$assignmentOutMulti, "_res.csv"))
-print(paste0(args$objectOutMulti,".rds"))
-print("params.csv")
+
 write.csv(hashtag$MULTI_ID, paste0(args$outputdir, "/", args$assignmentOutMulti, "_res.csv"))
 write.csv(params, paste0(args$outputdir, "/params.csv"))
 saveRDS(hashtag, paste0(args$outputdir, "/", args$objectOutMulti, ".rds"))

bin/bff.R

@@ -81,19 +81,41 @@ if(as.logical(args$preprocess)){
   counts <- Read10X(args$fileHto) 
 }
 
-if(args$methodsForConsensus=="bff_raw" || args$methodsForConsensus=="bff_cluster" || args$methodsForConsensus=="combined_bff" || is.null(args$methodsForConsensus)  )
+substring_vector <- NULL
+if (!is.null(args$methodsForConsensus)) { 
+  consensus_methods = args$methodsForConsensus
+  substring_vector <- strsplit(consensus_methods, ",")[[1]]
+}
+
+perCell_args <- args$perCellSaturation
+perCell <- ifelse(perCell_args == "null" || perCell_args == "Null", NULL, perCell_args)
+<<<<<<< HEAD
+<<<<<<< HEAD
+print("---------------------")
+print(perCell)
+print("---------------------")
+=======
+print(perCell)
+
+>>>>>>> c781241 (bff re-added problematic parameter)
+=======
+print("---------------------")
+print(perCell)
+print("---------------------")
+
+if(args$methodsForConsensus=="bff_raw" || args$methodsForConsensus=="bff_cluster" || args$methodsForConsensus=="bff_raw,bff_cluster" || is.null(args$methodsForConsensus)  )
   #Only Bff in its different variations is available
   if (args$methods == "bff_raw") {
     print("Executing BFF raw")
-    cell_hash_R_res <- GenerateCellHashingCalls(barcodeMatrix = counts, methods = c("bff_raw"), doTSNE = do_TSNE, doHeatmap = do_Heatmap, methodsForConsensus = args$methodsForConsensus,cellbarcodeWhitelist = args$cellbarcodeWhitelist, metricsFile= args$metricsFile, perCellSaturation= args$perCellSaturation, majorityConsensusThreshold = args$majorityConsensusThreshold, chemistry = args$chemistry, callerDisagreementThreshold = args$callerDisagreementThreshold )
+    cell_hash_R_res <- GenerateCellHashingCalls(barcodeMatrix = counts, methods = c("bff_raw"), doTSNE = do_TSNE, doHeatmap = do_Heatmap, methodsForConsensus = substring_vector,cellbarcodeWhitelist = args$cellbarcodeWhitelist, metricsFile = args$metricsFile, perCellSaturation = args$perCellSaturation, majorityConsensusThreshold = args$majorityConsensusThreshold, chemistry = args$chemistry, callerDisagreementThreshold = args$callerDisagreementThreshold )
   }else if (args$methods == "bff_cluster") {
     print("Executing BFF cluster")
-    cell_hash_R_res <- GenerateCellHashingCalls(barcodeMatrix = counts, methods = c("bff_cluster"), doTSNE = do_TSNE, doHeatmap = do_Heatmap)
-    #methodsForConsensus = args$methodsForConsensus,cellbarcodeWhitelist = args$cellbarcodeWhitelist, metricsFile= args$metricsFile, perCellSaturation= args$perCellSaturation, majorityConsensusThreshold = args$majorityConsensusThreshold, chemistry = args$chemistry, callerDisagreementThreshold = args$callerDisagreementThreshold 
+    cell_hash_R_res <- GenerateCellHashingCalls(barcodeMatrix = counts, methods = c("bff_cluster"), doTSNE = do_TSNE, doHeatmap = do_Heatmap,methodsForConsensus = substring_vector,cellbarcodeWhitelist = args$cellbarcodeWhitelist,metricsFile = args$metricsFile, perCellSaturation = args$perCellSaturation, majorityConsensusThreshold = args$majorityConsensusThreshold, chemistry = args$chemistry, callerDisagreementThreshold = args$callerDisagreementThreshold)
   }else if (args$methods == "combined_bff") {
     print("Executing BFF combined")
-    cell_hash_R_res <- GenerateCellHashingCalls(barcodeMatrix = counts, methods = c("bff_raw", "bff_cluster") , doTSNE = do_TSNE, doHeatmap = do_Heatmap )
-    #methodsForConsensus = args$methodsForConsensus,cellbarcodeWhitelist = args$cellbarcodeWhitelist, metricsFile= args$metricsFile, perCellSaturation= args$perCellSaturation, majorityConsensusThreshold = args$majorityConsensusThreshold, chemistry = args$chemistry, callerDisagreementThreshold = args$callerDisagreementThreshold
+    cell_hash_R_res <- GenerateCellHashingCalls(barcodeMatrix = counts, methods = c("bff_raw", "bff_cluster") , doTSNE = do_TSNE, doHeatmap = do_Heatmap,methodsForConsensus = substring_vector, cellbarcodeWhitelist = args$cellbarcodeWhitelist ,metricsFile = args$metricsFile, perCellSaturation = NULL, majorityConsensusThreshold = args$majorityConsensusThreshold )
+    #cell_hash_R_res <- GenerateCellHashingCalls(barcodeMatrix = counts, methods = c("bff_raw", "bff_cluster") , doTSNE = do_TSNE, doHeatmap = do_Heatmap,methodsForConsensus = substring_vector,cellbarcodeWhitelist = args$cellbarcodeWhitelist, metricsFile = args$metricsFile, perCellSaturation = NULL, majorityConsensusThreshold = args$majorityConsensusThreshold, chemistry = args$chemistry, callerDisagreementThreshold = args$callerDisagreementThreshold )
+
   }else {
     print("Method not available on the pipeline")
 }else{

bin/demuxem.py

@@ -2,8 +2,10 @@
 import pegasus as pg
 import demuxEM
 import numpy as np
+import scanpy as sc
 import argparse
 import pandas as pd
+import pegasusio as io
 
 parser = argparse.ArgumentParser(description='Parser for DemuxEM - Demultiplexing')
 parser.add_argument('--rna_matrix_dir', help= 'cellranger output folder which contains raw RNA count matrix in mtx format.')
@@ -29,14 +31,16 @@
 if __name__ == '__main__':
     output_name = args.outputdir + "/" + args.objectOutDemuxem
     # load input rna data
-    data = pg.read_input(args.rna_matrix_dir, modality="rna")
-    data.subset_data(modality_subset=['rna'])
-    data.concat_data() # in case of multi-organism mixing data
+    #data = io.read_input(args.rna_matrix_dir, modality="rna")
+    rna_data = sc.read_10x_mtx(args.rna_matrix_dir)
+    hashing_data = sc.read_10x_mtx(args.hto_matrix_dir,gex_only=False)
+    #data.subset_data(modality_subset=['rna'])
+    #data.concat_data() # in case of multi-organism mixing data
     # load input hashing data
-    data.update(pg.read_input(args.hto_matrix_dir, modality="hashing"))
+    #data.update(io.read_input(args.hto_matrix_dir, modality="hashing"))
     # Extract rna and hashing data
-    rna_data = data.get_data(modality="rna")
-    hashing_data = data.get_data(modality="hashing")
+    #rna_data = data.get_data(modality="rna")
+    #hashing_data = data.get_data(modality="hashing")
     filter = ""
     if args.filter_demuxem.lower() in ['true', 't', 'yes', 'y', '1']:
         filter = True
@@ -47,6 +51,7 @@
     # Filter the RNA matrix
     rna_data.obs["n_genes"] = rna_data.X.getnnz(axis=1)
     rna_data.obs["n_counts"] = rna_data.X.sum(axis=1).A1
+    #data.obs["n_counts"] = rna_data.X.sum(axis=1).A1
     if(filter):
         print("Filtering RNA matrix")
         obs_index = np.logical_and.reduce(
@@ -85,8 +90,10 @@
                 title="{gene_name}: a gender-specific gene".format(gene_name=gene_name),
             )
     # output results
+    mudata = io.MultimodalData(rna_data)
+    mudata.update(io.read_input(args.hto_matrix_dir, modality="hashing"))
     pg.write_output(demux_results, output_name + "_demux.zarr.zip")
-    pg.write_output(data, output_name + ".out.demuxEM.zarr.zip")
+    pg.write_output(mudata, output_name + ".out.demuxEM.zarr.zip")
     print("\nSummary statistics:")
     print("total\t{}".format(rna_data.shape[0]))
     for name, value in rna_data.obs["demux_type"].value_counts().iteritems():

bin/demuxmix.R

@@ -1,10 +1,4 @@
 #!/usr/bin/env Rscript
-if (!require("BiocManager", quietly = TRUE))
-    install.packages("BiocManager",repos = "http://cran.us.r-project.org")
-is_demuxmix<-require("demuxmix")
-if (!require("demuxmix", quietly = TRUE))
-    BiocManager::install("demuxmix")
-
 library(Seurat)
 library(demuxmix)
 library(argparse)
@@ -36,11 +30,10 @@ args <- parser$parse_args()
 if(as.logical(args$rna_available)){ 
 
     umi <- Read10X(data.dir = args$fileUmi, gene.column=args$gene_col, strip.suffix = TRUE)
-    #umi <- readRDS(args$fileUmi)
+
 }
 counts <- Read10X(data.dir = args$fileHto, gene.column=args$gene_col, strip.suffix = TRUE)
-#counts <- Read10X_h5(args$fileHto)
-#counts <- readRDS(args$fileHto)
+
 
 if(as.logical(args$rna_available)){
     Argument <- c("fileUmi", "fileHto")