Patch 2024 NMD skipping feature

cleaner.sh

@@ -2,8 +2,8 @@
 
 rm -r work/
 rm -r .nextflow/
-rm timeline-*
-rm trace-*
+rm timeline*
+rm trace*
 rm .nextflow.log*
 
 find . -maxdepth 4 -type d -name .idea -exec rm -r {} +

containers_build/boostdm/features/exon.py

@@ -6,51 +6,30 @@
 from boostdm.vepreader import Tabix
 
 
-def nmd_rule(exon, total_exons):
-
-    """
-    mutation is in an exon
-    if first or last exon then return 1
-    otherwise return 0
-    """
-
-    if exon == 0:
-        nmd = 0
-    elif (exon == 1) or (exon == total_exons):
-        nmd = 1
-    else:
-        nmd = 0
-    return nmd
-
-
-def get_exon(chr_, pos, alt,gene, reader):
+def get_nmd(chr_, pos, alt, gene, reader):
 
     for data in reader.get(chr_, pos, pos):
         alt_vep = (data["ALT"] == alt)
         mane_vep = (data["MANE_SELECT"] != '-') # impose mane transcript
         correct_gene = (data["SYMBOL"] == gene)  # skip cases with antisense overlapping gene
         if alt_vep and mane_vep and correct_gene:
-            exons = data["EXON"]
-            if '/' in exons:
-                exon, total_exons = tuple(exons.split('/'))
-            else:
-                exon, total_exons = 0, 0
-            return nmd_rule(exon, total_exons)
+            if data["NMD_SKIPPING"] == '-':
+                return 0
+            elif data["NMD_SKIPPING"] == 'NMD_escaping_variant':
+                return 1
     return 0
 
 
 def add_feature(df):
 
     df = df.copy()
     with Tabix(TABIX_FILE) as reader:
-        get_from_reader = partial(get_exon, reader=reader)
-        df['nmd'] = df.apply(lambda row: get_from_reader(str(row['chr']),
-                                                         int(row['pos']),
-                                                         row['alt'],
-                                                         row['gene']), axis=1)
+        get_from_reader = partial(get_nmd, reader=reader)
+        df['nmd'] = df.apply(lambda row: get_from_reader(row['chr'], row['pos'], row['alt'], row['gene']), axis=1)
     return df
 
 
+
 def test():
     """Test function"""
     df = pd.DataFrame({

containers_build/boostdm/vepreader.py

@@ -8,7 +8,7 @@
 HEADER = [    
     'CHR', 'POS', 'REF', 'ALT', 'GENE','ENST','TYPE','CNSQ','cDNA_POS',
     'CDS_POS', 'PROT_POS','AA','CODONS','EXISTING_VARIATION','IMPACT','DISTANCE','STRAND','FLAGS','SYMBOL',
-    'SYMBOL_SOURCE','HGNC_ID','CANONICAL','MANE_SELECT','MANE_PLUS_CLINICAL','ENSP','EXON','INTRON'
+    'SYMBOL_SOURCE','HGNC_ID','CANONICAL','MANE_SELECT','MANE_PLUS_CLINICAL','ENSP','EXON','INTRON','NMD_SKIPPING'
     ]
 
 

nextflow.config

@@ -7,10 +7,13 @@ params {
 env {
     GENOME_BUILD = "hg38"
     INTOGEN_DATASETS = "/workspace/datasets/intogen/runs/v2024/20240409_ALL/"
-    BOOSTDM_DATASETS = "/workspace/projects/intogen_plus/intogen-plus-v2024/datasets/boostdm/"
+
+    BOOSTDM_DATASETS = "/workspace/projects/intogen_plus/containers/datasets_patch_24_11/boostdm/"
+    // BOOSTDM_DATASETS = "/workspace/projects/intogen_plus/intogen-plus-v2024/datasets/boostdm/"
+
     VEP_SATURATION = env.INTOGEN_DATASETS + "/steps/boostDM/saturation/"
-    PIPELINE = "/workspace/datasets/boostdm_runs/boostdm-pipeline-2024/"
-    OUTPUT = "/workspace/datasets/boostdm_runs/boostdm-cancer-output-2024-noIARC/"
+    PIPELINE = "/workspace/datasets/boostdm_runs/boostdm-pipeline-patch-2024/"
+    OUTPUT = "/workspace/datasets/boostdm_runs/boostdm-patch-2024-output/"
     MAVE_DATA = "/workspace/projects/boostdm_analyses/mave_data/"
 }
 

scan_errors.sh

@@ -1,3 +1,7 @@
 # uses conda environment boostdm-new-pipeline
 
+# usage example: 
+
+# bash scan_errors.sh trace.txt
+
 cat $1 | grep FAILED | cut -f3 | python _scan_errors.py