scRNA-seq_analysis

pipelines/24_scanpy_to_seurat/scanpy_to_seurat.R

@@ -0,0 +1,98 @@
+args = commandArgs(trailingOnly=T)
+args = paste(args, collapse = "")
+args = unlist(strsplit(args, ";"))
+
+print("The scanpy object will be converted to a seurat object")
+print("Please keep in mind that the resulted seurat object will have some limitations.")
+print("Any dimensionality reduction computed for the scanpy object will not be transfered to the seurat object.")
+print("Furthermore the raw data will not be the original count data so you should be very very carefull subsetting or merging of the resulted seurat obj or just avoid doing these.")
+print("All seurat objects saved from this pipeline will have a tag appended saying 'converted_from_scanpy' to raise awarness about the processing history of the data.")
+
+arguments.list = "
+scanpy.addr.arg = args[1]
+save.to.arg     = args[2]
+"
+
+expected_arguments = unlist(strsplit(arguments.list, "\n"))
+expected_arguments = expected_arguments[!(expected_arguments == "")]
+
+if(length(args) != length(expected_arguments)){
+  error.msg = sprintf('This pipeline requires %s parameters', as.character(length(expected_arguments)))
+  expected_arguments = paste(unlist(lapply(strsplit(expected_arguments, ".arg"), "[", 1)), collapse = "\n")
+  stop(sprintf('This pipeline requires %s parameters: '))
+}
+
+eval(parse(text = arguments.list))
+
+for(n in 1:length(expected_arguments)){
+  argument = expected_arguments[n]
+  argument = gsub(pattern=" ", replacement="", x=argument)
+  argument.name = unlist(strsplit(argument, "="))[1]
+  variable.name = gsub(pattern=".arg", replacement="", argument.name)
+  argument.content = eval(parse(text = argument.name))
+  eval(parse(text = argument.content))
+  if (!exists(variable.name)){
+    stop(sprintf("Argument %s not passed. Stopping ... ", variable.name))
+  }
+}
+
+# create required folders for output and work material
+output_folder = "scanpy_to_seurat_"
+c.time = Sys.time()
+c.time = gsub(pattern=" BST", replacement="", x=c.time)
+c.time = gsub(pattern=":", replacement="", x=c.time)
+c.time = gsub(pattern=" ", replacement="", x=c.time)
+c.time = gsub(pattern="-", replacement="", x=c.time)
+c.time = substr(x=c.time, start=3, stop=nchar(c.time))
+output_folder = paste(output_folder, c.time, sep = "")
+output_folder = file.path("../../output", output_folder)
+output_folder = paste(output_folder, paste(sample(c(letters, LETTERS), 10, F), collapse = ""), sep = "_")
+dir.create(output_folder)
+
+scanpy.addr = file.path("../../data", scanpy.addr)
+save.to = paste(unlist(strsplit(save.to, "\\.RDS")), "converted_from_scanpy.RDS", sep = "_")
+
+source("../../tools/bunddle_utils.R")
+
+library(Seurat)
+library(methods)
+library(Matrix)
+
+#######################################################################################################
+
+# load data
+command = sprintf("%s scanpy_to_seurat.py %s %s", python.addr, scanpy.addr, output_folder)
+print("deconstructing the scanpy object in Python ...")
+system(command, wait = T)
+
+# load gene names
+print("Loading gene names ...")
+gene.names = read.csv(file.path(output_folder, "gene_names.csv"), header = F, row.names=1, stringsAsFactors = F)$V2
+
+# load cell names
+print("Loading cell names ...")
+cell.names = read.csv(file.path(output_folder, "cell_names.csv"), header = F, row.names=1, stringsAsFactors = F)$V2
+
+# load meta data
+print("Loading meta data ... ")
+meta.data = read.csv(file.path(output_folder, "meta_data.csv"), stringsAsFactors = F)
+
+# load the expression matrix
+print("Loading the expression matrix ...")
+expression.data = readMM(file.path(output_folder, 'expression.mtx'))
+expression.data = t(expression.data)
+
+# make the seurat object
+print("Constructing the Seurat object")
+colnames(expression.data) = cell.names
+rownames(expression.data) = gene.names
+seurat.obj = CreateSeuratObject(raw.data = expression.data, min.cells = 0, min.genes = 0, project = "")
+seurat.obj@meta.data = cbind(seurat.obj@meta.data, meta.data)
+seurat.obj@data = expression.data
+
+print("Saving the seurat object ... ")
+saveRDS(seurat.obj, save.to)
+
+unlink(output_folder, recursive=T, force=T)
+
+print("Ended beautifully ... ")

pipelines/24_scanpy_to_seurat/scanpy_to_seurat.py

@@ -0,0 +1,48 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""
+Created on Thu Dec 20 12:00:47 2018
+
+@author: doru
+"""
+
+# gene expression should be at data.X
+# gene names should be at data.var_names
+# cell names should be in data.obs_names and must be unique
+# categories (the meta data) should be columns at data.obs
+# any exception from the above assumptions will result in errors
+
+import sys
+file_name     = sys.argv[1]
+output_folder = sys.argv[2]
+print("printing file_name")
+print(file_name)
+print("printing output folder")
+print(output_folder)
+
+import matplotlib; matplotlib.use('Agg');
+import scanpy.api as sc
+from os.path import join
+from scipy.io import mmwrite
+
+data = sc.read(file_name)
+print("printing data")
+print(data)
+
+# save expression data
+expression_data_filename = join(output_folder, 'expression')
+gene_expression = data.X
+mmwrite(expression_data_filename, gene_expression)
+
+# save gene names
+gene_names_file = join(output_folder, "gene_names.csv")
+data.var_names.to_series().to_csv(gene_names_file)
+
+# save cell names
+cell_names_file = join(output_folder, "cell_names.csv")
+data.obs_names.to_series().to_csv(cell_names_file)
+ 
+# get categories
+meta_data = join(output_folder, "meta_data.csv")
+data.obs.to_csv(meta_data)
+

pipelines/24_scanpy_to_seurat/scanpy_to_seurat.sh

@@ -0,0 +1,11 @@
+#!/bin/bash
+
+#$ -cwd
+#$ -N scanpy_to_seurat
+#$ -V
+#$ -l h_rt=23:59:59
+#$ -l h_vmem=100G
+
+Rscript scanpy_to_seurat.R $1
+
+echo "End on `date`"