Various code improvements

gene_splicer/utils.py

@@ -12,6 +12,8 @@
 import glob
 from pathlib import Path
 from csv import DictWriter, DictReader
+from itertools import groupby
+from operator import itemgetter
 
 logger = logging.getLogger(__name__)
 
@@ -412,39 +414,26 @@ def align(target_seq,
     ]
 
 def iterate_hivintact_data(name, outpath):
-    intact = {}
+    intact = set()
 
     def get_verdict(SEQID, all_errors):
         ordered = sorted(all_errors, key=HIVINTACT_ERRORS_TABLE.index)
         verdict = ordered[0]
         return [SEQID, verdict]
 
-    for d in glob.glob(str(outpath / 'hivintact*')):
+    for d in outpath.glob('hivintact*'):
         for (SEQID, sequence) in read_fasta(os.path.join(d, 'intact.fasta')):
-            row = [SEQID, 'Intact']
-            intact[SEQID] = True
-            yield row
+            yield [SEQID, 'Intact']
+            intact.add(SEQID)
 
-        sequence_name = None
         with open(os.path.join(d, 'errors.csv'), 'r') as f:
             reader = csv.DictReader(f)
+            grouped = groupby(reader, key=itemgetter('sequence_name'))
+            for sequence_name, errors in grouped:
+                if sequence_name not in intact:
+                    all_errors = [error['error'] for error in errors]
+                    yield get_verdict(sequence_name, all_errors)
 
-            last_name = None
-            all_errors = []
-            for row in reader:
-                sequence_name = row['sequence_name']
-                if sequence_name in intact: continue
-
-                if last_name != sequence_name and last_name is not None:
-                    if all_errors:
-                        yield get_verdict(last_name, all_errors)
-                        all_errors = []
-
-                all_errors.append(row['error'])
-                last_name = sequence_name
-
-        if all_errors:
-            yield get_verdict(sequence_name, all_errors)
 
 def get_hivintact_data(name, outpath):
     column_names = ['SEQID', 'MyVerdict']
@@ -572,16 +561,14 @@ def generate_proviral_landscape_csv(outpath, is_hivintact):
     landscape_rows = []
 
     table_precursor_csv = os.path.join(outpath, 'table_precursor.csv')
-    blastn_csv = glob.glob(
-        os.path.join(outpath, 'hivintact*', 'blast.csv') \
-        if is_hivintact else \
-        os.path.join(
-            outpath,
-            'hivseqinr*',
-            'Results_Intermediate',
-            'Output_Blastn_HXB2MEGA28_tabdelim.txt'
-        )
-    )[0]
+
+    if is_hivintact:
+        subpath = os.path.join(outpath, 'hivintact*', 'blast.csv')
+    else:
+        subpath = os.path.join(outpath, 'hivseqinr*', 'Results_Intermediate', 'Output_Blastn_HXB2MEGA28_tabdelim.txt')
+
+    blastn_csvs = glob.glob(subpath)
+    blastn_csv = blastn_csvs[0]
 
     with open(blastn_csv, 'r') as blastn_file:
         if is_hivintact: