Upload New File

notebook/scVAR_workflow_example.ipynb

+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#Import all dependencies from the scVAR enviroment (see installation instructions)\n",
+    "#from scVAR import *\n",
+    "import sys\n",
+    "import pickle\n",
+    "import os\n",
+    "import scanpy as sc\n",
+    "import pandas as pd"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#Check if all the arguments are in place \n",
+    "if len(sys.argv) < 2:\n",
+    "    print(\"Errore: specify sample name as argument.\")\n",
+    "    sys.exit(1)\n",
+    "\n",
+    "sample = sys.argv[1]\n",
+    "out_path = '/CNRITB/lcelli/SCVAR/BLL_August/output/' + sample + '/'\n",
+    "in_path = '/CNRITB/lcelli/SCVAR/BLL_August/input/' + sample + '/'\n",
+    "\n",
+    "# Crea\n",
+    "if not os.path.exists(out_path):\n",
+    "    os.makedirs(out_path, exist_ok=True)\n",
+    "\n",
+    "print('Start Analysis', sample)\n",
+    "\n",
+    "# Specify transcriptomics file path \n",
+    "tra_mat = in_path + 'matrix.mtx'\n",
+    "barcode_tra = in_path + 'clean_barcodes.txt'\n",
+    "feature = in_path + 'features.tsv'\n",
+    "# Specify genomic file path\n",
+    "var_mat = in_path + 'consensus_filtered_markdup.mtx'\n",
+    "barcode_var = in_path + 'barcodes_var.tsv'\n",
+    "snv = in_path + 'variants_filtered_markdup.txt'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Analize trascritomics and genomics separately\n",
+    "adata = transcriptomicAnalysis(matrix_path=tra_mat, bcode_path=barcode_tra, feature_path=feature, bcode_variants=barcode_var)\n",
+    "adata = variantAnalysis(adata, matrix_path=var_mat, bcode_path=barcode_var, variants_path=snv)\n",
+    "\n",
+    "#Perform data integration\n",
+    "adata = omicsIntegration(adata)\n",
+    " \n",
+    "# Compute transcriptomics, genomics and integrated clusters at different resolutions\n",
+    "for res in [0.01, 0.05, 0.5]:\n",
+    "    adata = calcOmicsClusters(adata, omic_key='variant', res=res)\n",
+    "    adata = calcOmicsClusters(adata, omic_key='trans', res=res)\n",
+    "    adata = calcOmicsClusters(adata, omic_key='int', res=res)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Optionally add metadata from other scRNA-seq analysis tools (i.e., Seurat)\n",
+    "md = pd.read_csv('/CNRITB/lcelli/SCVAR/BLL_August/input/' + sample + '/metadata.csv', header=0, index_col=0)\n",
+    "md = md.loc[list(adata.obs.index)]\n",
+    "for metadata in ['SingleR_DatabaseImmuneCellExpressionData_labels', 'orig.ident', 'Phase', 'RNA_snn_res.0.6']:\n",
+    "    adata.obs[metadata] = md[metadata]\n",
+    "adata.obs['RNA_snn_res.0.6'] = adata.obs['RNA_snn_res.0.6'].astype(str)\n",
+    "\n",
+    "# Save all data \n",
+    "with open(out_path + sample + '_adata.pkl', 'wb') as f:\n",
+    "    pickle.dump(adata, f)\n",
+    "\n",
+    "#####Perform preliminary plotting#####\n",
+    "\n",
+    "sc.set_figure_params(scanpy=True, dpi=80, dpi_save=150, frameon=True, vector_friendly=True, fontsize=14, figsize=(12,8), color_map=None, format='png', facecolor='#FFFFFF', transparent=False, ipython_format='png2x')\n",
+    "\n",
+    "# Plotting\n",
+    "for res in [0.01, 0.05, 0.5]:\n",
+    "    for omic in ['variant', 'trans', 'int']:\n",
+    "        p = sc.pl.embedding(adata, basis='int_umap', color=omic + '_clust_' + str(res), title=[sample + ' UMAP:' + 'int' + ' Cluster:' + omic + ' res' + str(res)], size=10, frameon=False, return_fig=True)\n",
+    "        p.savefig(out_path + sample + '_INT_umap_' + omic + '_cluster_res' + str(res) + '.png')\n",
+    "\n",
+    "for mtdt in ['SingleR_DatabaseImmuneCellExpressionData_labels', 'orig.ident', 'Phase', 'RNA_snn_res.0.6']:\n",
+    "    p = sc.pl.embedding(adata, basis='int_umap', color=mtdt, title=[sample + ' UMAP:' + 'int ' + mtdt], size=10, frameon=False, return_fig=True)\n",
+    "    p.savefig(out_path + sample + '_INT_umap_' + mtdt + '.png')\n"
+   ]
+  }
+ ],
+ "metadata": {
+  "celltoolbar": "Edit Metadata",
+  "kernelspec": {
+   "display_name": "Python [conda env:smile]",
+   "language": "python",
+   "name": "conda-env-smile-py"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.2"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}