changeset 6:0b6a51c7e008 draft

planemo upload for repository https://github.com/peterjc/pico_galaxy/tree/master/tools/predictnls commit 4bd49529e9ca2096cd875e98daf7190d13fa8d0b-dirty

--- a/tools/predictnls/README.rst	Wed Aug 05 12:24:56 2015 -0400
+++ b/tools/predictnls/README.rst	Wed Feb 01 13:13:25 2017 -0500
@@ -79,12 +79,12 @@
 Planemo commands (which requires you have set your Tool Shed access details in
 ``~/.planemo.yml`` and that you have access rights on the Tool Shed)::
 
-    $ planemo shed_update --shed_target testtoolshed --check_diff ~/repositories/pico_galaxy/tools/predictnls/
+    $ planemo shed_update -t testtoolshed --check_diff ~/repositories/pico_galaxy/tools/predictnls/
     ...
 
 or::
 
-    $ planemo shed_update --shed_target toolshed --check_diff ~/repositories/pico_galaxy/tools/predictnls/
+    $ planemo shed_update -t toolshed --check_diff ~/repositories/pico_galaxy/tools/predictnls/
     ...
 
 To just build and check the tar ball, use::

--- a/tools/predictnls/predictnls.py	Wed Aug 05 12:24:56 2015 -0400
+++ b/tools/predictnls/predictnls.py	Wed Feb 01 13:13:25 2017 -0500
@@ -1,13 +1,13 @@
 #!/usr/bin/env python
 
-#Copyright 2011-2013 by Peter Cock, James Hutton Institute (formerly SCRI), UK
+# Copyright 2011-2013 by Peter Cock, James Hutton Institute (formerly SCRI), UK
 #
-#Licenced under the GPL (GNU General Public Licence) version 3.
+# Licenced under the GPL (GNU General Public Licence) version 3.
 #
-#Based on Perl script predictNLS v1.3, copyright 2001-2005 and the later
-#versions up to predictnls v1.0.20 (copright 2012), by Rajesh Nair
-#([e-mail]) and Burkhard Rost ([e-mail]), Rost Lab,
-#Columbia University http://rostlab.org/
+# Based on Perl script predictNLS v1.3, copyright 2001-2005 and the later
+# versions up to predictnls v1.0.20 (copright 2012), by Rajesh Nair
+# ([e-mail]) and Burkhard Rost ([e-mail]), Rost Lab,
+# Columbia University http://rostlab.org/
 
 """Batch mode predictNLS, for finding nuclear localization signals
 
@@ -53,30 +53,27 @@
 import sys
 import re
 
-def sys_exit(msg, return_code=1):
-    sys.stderr.write(msg.rstrip() + "\n")
-    sys.exit(return_code)
-
 if len(sys.argv) == 4:
     fasta_filename, tabular_filename, re_filename = sys.argv[1:]
 elif len(sys.argv) == 3:
     fasta_filename, tabular_filename = sys.argv[1:]
-    #Use os.path.realpath(...) to handle being called via a symlink
-    #Try under subdirectory data:
+    # Use os.path.realpath(...) to handle being called via a symlink
+    # Try under subdirectory data:
     re_filename = os.path.join(os.path.dirname(os.path.realpath(sys.argv[0])),
                                "data", "My_NLS_list")
     if not os.path.isfile(re_filename):
-        #Try in same directory as this script:
+        # Try in same directory as this script:
         re_filename = os.path.join(os.path.dirname(os.path.realpath(sys.argv[0])),
                                                    "My_NLS_list")
 else:
-    sys_exit("Expect 2 or 3 arguments: input FASTA file, output tabular file, and NLS motif file")
+    sys.exit("Expect 2 or 3 arguments: input FASTA file, output tabular file, and NLS motif file")
 
 if not os.path.isfile(fasta_filename):
-    sys_exit("Could not find FASTA input file: %s" % fasta_filename)
+    sys.exit("Could not find FASTA input file: %s" % fasta_filename)
 
 if not os.path.isfile(re_filename):
-    sys_exit("Could not find NLS motif file: %s" % re_filename)
+    sys.exit("Could not find NLS motif file: %s" % re_filename)
+
 
 def load_re(filename):
     """Parse the 5+ column tabular NLS motif file."""
@@ -92,7 +89,7 @@
             regex = re.compile(regex)
             p_count = int(p_count)
         except ValueError:
-            sys_exit("Bad data in line: %s" % line)
+            sys.exit("Bad data in line: %s" % line)
         if 6 <= len(parts):
             proteins = parts[5]
             assert p_count == len(proteins.split(",")), line
@@ -105,10 +102,11 @@
         else:
             domains = ""
             assert p_count == 0
-        #There can be further columns (DNA binding?), but we don't use them.
+        # There can be further columns (DNA binding?), but we don't use them.
         yield regex, evidence, p_count, percent_nuc, proteins, domains
     handle.close()
 
+
 def fasta_iterator(filename):
     """Simple FASTA parser yielding tuples of (name, upper case sequence)."""
     if filename == "-":
@@ -120,15 +118,15 @@
         if line.startswith(">"):
             if name:
                 yield name, seq
-            #Take the first word only as the name:
-            name = line[1:].rstrip().split(None,1)[0]
+            # Take the first word only as the name:
+            name = line[1:].rstrip().split(None, 1)[0]
             seq = ""
         elif name:
-            #Simple way would leave in any internal white space,
-            #seq += line.strip().upper()
+            # Simple way would leave in any internal white space,
+            # seq += line.strip().upper()
             seq += "".join(line.strip().upper().split())
         elif not line.strip():
-            #Ignore blank lines before first record
+            # Ignore blank lines before first record
             pass
         else:
             raise ValueError("Bad FASTA line %r" % line)
@@ -150,14 +148,14 @@
 nls = 0
 for idn, seq in fasta_iterator(fasta_filename):
     for regex, evidence, p_count, percent_nuc_prot, proteins, domains in motifs:
-        #Perl predictnls v1.0.17 (and older) take right most hit only, Bug #40
-        #This has been fixed (v1.0.18 onwards, June 2011), so we return all the matches
+        # Perl predictnls v1.0.17 (and older) take right most hit only, Bug #40
+        # This has been fixed (v1.0.18 onwards, June 2011), so we return all the matches
         for match in regex.finditer(seq):
-            #Perl predictnls v1.0.17 (and older) return NLS start position with zero
-            #but changed to one based counting in v1.0.18 (June 2011) onwards, Bug #38
-            #We therefore also use one based couting, hence the start+1 here:
-            out_handle.write("%s\t%i\t%s\t%s\t%s\t%i\t%s\t%s\t%s\n" \
-                             % (idn, match.start()+1, match.group(),
+            # Perl predictnls v1.0.17 (and older) return NLS start position with zero
+            # but changed to one based counting in v1.0.18 (June 2011) onwards, Bug #38
+            # We therefore also use one based couting, hence the start+1 here:
+            out_handle.write("%s\t%i\t%s\t%s\t%s\t%i\t%s\t%s\t%s\n"
+                             % (idn, match.start() + 1, match.group(),
                                 regex.pattern, evidence, p_count,
                                 percent_nuc_prot, proteins, domains))
             nls += 1