Fix annotation processing - closes #93

aertslab · Jan 31, 2023 · 4416843 · 4416843
1 parent 8e3c0df
commit 4416843
Showing 1 changed file with 5 additions and 5 deletions.
diff --git a/notebooks/pbmc_multiome_tutorial.ipynb b/notebooks/pbmc_multiome_tutorial.ipynb
@@ -1015,12 +1015,12 @@
     "import pybiomart as pbm\n",
     "dataset = pbm.Dataset(name='hsapiens_gene_ensembl',  host='http://www.ensembl.org')\n",
     "annot = dataset.query(attributes=['chromosome_name', 'transcription_start_site', 'strand', 'external_gene_name', 'transcript_biotype'])\n",
-    "annot['Chromosome/scaffold name'] = annot['Chromosome/scaffold name'].to_numpy(dtype = str)\n",
-    "filter = annot['Chromosome/scaffold name'].str.contains('CHR|GL|JH|MT')\n",
-    "annot = annot[~filter]\n",
-    "annot['Chromosome/scaffold name'] = annot['Chromosome/scaffold name'].str.replace(r'(\\b\\S)', r'chr\\1')\n",
     "annot.columns=['Chromosome', 'Start', 'Strand', 'Gene', 'Transcript_type']\n",
-    "annot = annot[annot.Transcript_type == 'protein_coding']\n",
+    "annot['Chromosome'] = annot['Chromosome'].astype(str)\n",
+    "mask = ~annot['Chromosome'].str.contains('CHR|GL|JH|MT') & (annot['Transcript_type'] == 'protein_coding')\n",
+    "annot = annot[mask].reset_index(drop=True)\n",
+    "annot['Chromosome'] = annot['Chromosome'].apply(lambda x: f\"chr{x}\" if not x.startswith(\"KI\") else x)\n",
+    "\n",
     "from pycisTopic.qc import *\n",
     "path_to_regions = {'10x_pbmc':os.path.join(work_dir, 'scATAC/consensus_peak_calling/consensus_regions.bed')}\n",
     "\n",