Copy over deeptoolsintervals fixes

Author: dpryan79

CHANGES.txt

@@ -1,3 +1,7 @@
+2.3.3
+
+ * The deeptoolsinterval module was modified to speed up plotEnrichment, which was taking forever to finish.
+
 2.3.1
 
  * This release has no real code changes, the 2.3.0 release on pypi was missing files.

deeptools/_version.py

@@ -2,4 +2,4 @@
 # This file is originally generated from Git information by running 'setup.py
 # version'. Distribution tarballs contain a pre-generated copy of this file.
 
-__version__ = '2.3.1'
+__version__ = '2.3.2'

deeptoolsintervals/enrichment.py

@@ -4,7 +4,7 @@
 from deeptoolsintervals.parse import GTF, openPossiblyCompressed
 import sys
 from os.path import basename
-import shlex
+import csv
 
 
 class Enrichment(GTF):
@@ -110,9 +110,10 @@ def parseGTF(self, fp, line):
             strand = 1
 
         feature = cols[2]
-        s = shlex.split(cols[8])
-        if "deepTools_group" in s and s[-1] != "deepTools_group":
-            feature = s[s.index("deepTools_group") + 1].rstrip(";")
+        if "deepTools_group" in cols[8]:
+            s = next(csv.reader([cols[8]], delimiter=' '))
+            if s[-1] != "deepTools_group":
+                feature = s[s.index("deepTools_group") + 1].rstrip(";")
 
         self.tree.addEnrichmentEntry(self.mungeChromosome(cols[0]), int(cols[3]) - 1, int(cols[4]), strand, cols[5], feature)
         if feature not in self.features:
@@ -134,9 +135,10 @@ def parseGTF(self, fp, line):
                     strand = 1
 
                 feature = cols[2]
-                s = shlex.split(cols[8])
-                if "deepTools_group" in s and s[-1] != "deepTools_group":
-                    feature = s[s.index("deepTools_group") + 1].rstrip(";")
+                if "deepTools_group" in cols[8]:
+                    s = next(csv.reader([cols[8]], delimiter=" "))
+                    if s[-1] != "deepTools_group":
+                        feature = s[s.index("deepTools_group") + 1].rstrip(";")
 
                 self.tree.addEnrichmentEntry(self.mungeChromosome(cols[0]), int(cols[3]) - 1, int(cols[4]), strand, cols[5], feature)
                 if feature not in self.features:

deeptoolsintervals/parse.py

@@ -9,7 +9,7 @@
 except:
     supportsBZ2 = False
 import os.path
-import shlex
+import csv
 
 
 def getNext(fp):
@@ -41,7 +41,7 @@ def seemsLikeGTF(cols):
         cols[6] in ['+', '-', '.']
         if cols[7] != '.':
             int(cols[7]) in [0, 1, 2]
-        s = shlex.split(cols[8])
+        s = next(csv.reader([cols[8]], delimiter=' '))
         assert("gene_id" in s)
         assert(s[-1] != "gene_id")
         return True
@@ -373,7 +373,7 @@ def parseGTFtranscript(self, cols, label):
             sys.stderr.write("Warning: non-GTF line encountered! {0}\n".format("\t".join(cols)))
             return
 
-        s = shlex.split(cols[8])
+        s = next(csv.reader([cols[8]], delimiter=' '))
         if "deepTools_group" in s and s[-1] != "deepTools_group":
             label = s[s.index("deepTools_group") + 1].rstrip(";")
         elif self.defaultGroup is not None:
@@ -420,7 +420,7 @@ def parseGTFexon(self, cols):
             sys.stderr.write("Warning: Invalid start in '{0}', skipping\n".format("\t".join(cols)))
             return
 
-        s = shlex.split(cols[8])
+        s = next(csv.reader([cols[8]], delimiter=' '))
         if self.transcript_id_designator not in s or s[-1] == self.transcript_id_designator:
             sys.stderr.write("Warning: {0} is malformed!\n".format("\t".join(cols)))
             return