Add new profiles to hadge

.github/workflows/test_action.yml

@@ -0,0 +1,22 @@
+name: hadge test workflow
+on: push
+jobs:
+  test:
+    name: Run pipeline with test data
+    runs-on: ubuntu-latest
+    steps:
+      - name: Check out pipeline code
+        uses: actions/checkout@v3
+      - name: Setup conda
+        uses: conda-incubator/setup-miniconda@v2
+        with:
+          auto-update-conda: true
+          miniconda-version: "latest"
+          channels: conda-forge, bioconda
+      - name: Install Nextflow
+        uses: nf-core/setup-nextflow@v1
+      - name: Download test dataset
+        run: bash ${GITHUB_WORKSPACE}/test_data/download_data.sh
+      - name: Run pipeline with test data
+        run: |
+          nextflow run ${GITHUB_WORKSPACE} -profile test,conda --souporcell False --generate_anndata False

bin/HTODemux-visualisation.R

@@ -9,30 +9,30 @@ parser <- ArgumentParser("Parameters for HTODemux Visualisation")
 parser$add_argument("--hashtagPath",help="folder where rds object was saved from the first part of HTODemux")
 parser$add_argument("--assay",help="Name of the Hashtag assay HTO by default", default = "HTO")
 #Output graphs - Ridge Plot
-parser$add_argument("--ridgePlot", help = "Generates a ridge plot from the results, True to generate", default = "TRUE")
+parser$add_argument("--ridgePlot", help = "Generates a ridge plot from the results, True to generate", default = "True")
 parser$add_argument("--ridgeNCol", help = "Number of columns for ridgePlot", default = 3, type = "integer")
 
 #Output graphs - Scatter Feature
-parser$add_argument("--featureScatter",help = "Generates a ridge plot from the results, True to generate", default = "TRUE")
+parser$add_argument("--featureScatter",help = "Generates a ridge plot from the results, True to generate", default = "True")
 parser$add_argument("--scatterFeat1", help = "Feature 1 for Feature Scatter Plot", default = "hto_HTO-A")
 parser$add_argument("--scatterFeat2", help = "Feature 2 for Feature Scatter Plot", default = "hto_HTO-B")
 
 #Output graphs - Violin Plot
-parser$add_argument("--vlnPlot", help = "Generates a violin plot from the results, True to generate", default = "TRUE")
+parser$add_argument("--vlnPlot", help = "Generates a violin plot from the results, True to generate", default = "True")
 parser$add_argument("--vlnFeatures", help = "Features to plot (gene expression, metrics, PC scores, anything that can be retreived by FetchData)", default = "nCount_RNA")
 parser$add_argument("--vlnLog", help = "plot the feature axis on log scale", action = "store_true")
 
 #Output graphs - tSNE
-parser$add_argument("--tSNE", help = "Generate a two dimensional tSNE embedding for HTOs", default = "TRUE")
+parser$add_argument("--tSNE", help = "Generate a two dimensional tSNE embedding for HTOs", default = "True")
 parser$add_argument("--tSNEIdents", help = "What should we remove from the object (we have Singlet,Doublet and Negative)", default = "Negative")
-parser$add_argument("--tSNEInvert", action = "store_true") # TRUE
-parser$add_argument("--tSNEVerbose", action = "store_true") # FALSE
-parser$add_argument("--tSNEApprox", action = "store_true") # FALSE
+parser$add_argument("--tSNEInvert", action = "store_true")
+parser$add_argument("--tSNEVerbose", action = "store_true")
+parser$add_argument("--tSNEApprox", action = "store_true")
 parser$add_argument("--tSNEDimMax", help = "max number of donors ",type = "integer", default = 1)
 parser$add_argument("--tSNEPerplexity", help = "value for perplexity", type = "integer",  default = 100)
 
 #Output graphs - Heatmap
-parser$add_argument("--heatMap", help = "Generate a Heatmap", default = "FALSE")
+parser$add_argument("--heatMap", help = "Generate a Heatmap", default = "False")
 parser$add_argument("--heatMapNcells", help ="value for number of cells", type = "integer",  default = 500)
 parser$add_argument("--outputdir", help='Output directory')
 
@@ -56,24 +56,24 @@ hashtag <-readRDS(hash_file)
 
 # Ridge Plot
 # Group cells based on the max HTO signal
-if (args$ridgePlot == "TRUE") {
+if (args$ridgePlot == "True") {
   Idents(hashtag) <- paste0(args$assay, "_maxID")
   RidgePlot(hashtag, assay = args$assay, features = rownames(hashtag[[args$assay]]), ncol = args$ridgeNCol)
   ggsave(paste0(args$outputdir, '/ridge.jpeg'), device = 'jpeg', dpi = 500) # height = 10, width = 10
 }
 
-if (args$featureScatter == "TRUE") {
+if (args$featureScatter == "True") {
   FeatureScatter(hashtag, feature1 = args$scatterFeat1, feature2 = args$scatterFeat2)
   ggsave(paste0(args$outputdir, '/featureScatter.jpeg'), device = 'jpeg',dpi = 500)
 }
 
-if (args$vlnPlot == "TRUE") {
+if (args$vlnPlot == "True") {
   Idents(hashtag) <- paste0(args$assay, "_classification.global")
   VlnPlot(hashtag, features = args$vlnFeatures, pt.size = 0.1, log = args$vlnLog)
   ggsave(paste0(args$outputdir, '/violinPlot.jpeg'), device = 'jpeg', dpi = 500)
 }
 
-if (args$tSNE == "TRUE") {
+if (args$tSNE == "True") {
   hashtag.subset <- subset(hashtag, idents = args$tSNEIdents, invert = args$tSNEInvert)
   DefaultAssay(hashtag.subset) <- args$assay
   hashtag.subset <- ScaleData(hashtag.subset, features = rownames(hashtag.subset),
@@ -84,7 +84,7 @@ if (args$tSNE == "TRUE") {
   ggsave(paste0(args$outputdir, '/tSNE.jpeg'), device = 'jpeg', dpi = 500)
 }
 
-if (args$heatMap == "TRUE") {
+if (args$heatMap == "True") {
   HTOHeatmap(hashtag, assay = args$assay, ncells = args$heatMapNcells)
   ggsave(paste0(args$outputdir, '/heatMap.jpeg'), device = 'jpeg', dpi = 500)
 }

bin/demuxem.py

@@ -18,7 +18,7 @@
 parser.add_argument('--alpha_noise', help='The Dirichlet prior concenration parameter on the background noise.', type=float, default=1.0)
 parser.add_argument('--tol', help='Threshold used for the EM convergence.', type=float, default=1e-6)
 parser.add_argument('--n_threads', help='Number of threads to use. Must be a positive integer.', type=int, default=1)
-parser.add_argument('--filter_demuxem', help='Use the filter for RNA, true or false', default='true')
+parser.add_argument('--filter_demuxem', help='Use the filter for RNA, True or False', default='True')
 parser.add_argument('--generateGenderPlot', help='Generate violin plots using gender-specific genes (e.g. Xist). <gene> is a comma-separated list of gene names.', default='')
 parser.add_argument('--objectOutDemuxem', help='Output name of demultiplexing results. All outputs will use it as the prefix.', default="demuxem_res")
 parser.add_argument('--outputdir', help='Output directory')
@@ -31,16 +31,8 @@
 if __name__ == '__main__':
     output_name = args.outputdir + "/" + args.objectOutDemuxem
     # load input rna 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(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")
     filter = ""
     if args.filter_demuxem.lower() in ['true', 't', 'yes', 'y', '1']:
         filter = True
@@ -96,7 +88,7 @@
     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():
+    for name, value in rna_data.obs["demux_type"].value_counts().items():
         print("{}\t{}".format(name, value))
     summary = rna_data.obs["demux_type"].value_counts().rename_axis('classification').reset_index(name='counts')
     total = ["total", rna_data.shape[0]]

bin/generate_data.py

@@ -3,7 +3,6 @@
 import os
 import scanpy as sc
 import argparse
-import muon as mu
 
 parser = argparse.ArgumentParser(description="Parameters for generating anndata and mudata")
 parser.add_argument("--assignment", help="Folder which contains cSV file with demultiplexing assignment", default=None)