Merge pull request #45 from yhoogstrate/manual_things

Better way of handling stdio / logging for bioconda compatibility

Changelog

@@ -1,3 +1,6 @@
+2017-03-02	Youri Hoogstrate	v0.3.4
+	* Better way of handling stdio / logging for bioconda compatibility
+
 2017-02-08	Youri Hoogstrate	v0.3.3
 	* Huge improvement in performance in `extract subnetworks` by translating
 	  a recursion problem into an iterative linear problem.

bin/dr-disco

@@ -4,8 +4,6 @@
 
 
 import drdisco
-import logging
-import sys
 import pysam
 import click
 
@@ -41,9 +39,6 @@ from drdisco.ChimericAlignment import ChimericAlignment
 """
 
 
-logging.basicConfig(level=logging.DEBUG, format=drdisco.__log_format__, stream=sys.stdout)
-
-
 def main():
     CLI()
 

drdisco/ChimericAlignment.py

@@ -3,14 +3,14 @@
 # vim: set expandtab tabstop=4 shiftwidth=4 softtabstop=4 textwidth=79:
 
 import os
-import logging
 
 import pysam
 
 from fuma.Fusion import STRAND_FORWARD
 from .CigarAlignment import CigarAlignment
 
 from drdisco import __version__
+from drdisco import log
 
 """[License: GNU General Public License v3 (GPLv3)]
 
@@ -282,7 +282,7 @@ def fix_chain(self, alignments, bam_file, mates):
                 # start = self.get_closest(next_pos, alignments)
 
             # else:
-                # print("Warning - mates do not correspond? - maybe empty (-1) as well?")
+                # logger.warn("Warning - mates do not correspond? - maybe empty (-1) as well?")
 
                 # next_pos = [alignments[0].reference_id, alignments[0].reference_start]
                 # last_pos = [mates[0].reference_id, mates[0].reference_start]
@@ -459,7 +459,7 @@ def reconstruct_alignments(self, alignments, bam_file, fh_out):
 
         else:
             if n == 1:
-                print("Warning: segments of mate are missing: " + alignments[0].query_name)
+                log.warn("segments of mate are missing: " + alignments[0].query_name)
                 all_reads_updated.append(alignments[0])
             else:
                 raise Exception("what happens here?")
@@ -480,10 +480,10 @@ def convert(self, bam_file_discordant_fixed, temp_dir):
         # @TODO / consider todo - start straight from sam
         # samtools view -bS samples/7046-004-041_discordant.Chimeric.out.sam > samples/7046-004-041_discordant.Chimeric.out.unsorted.bam
 
-        logging.info("Convert into a name-sorted bam file, to get all reads with the same name adjacent to each other")
+        log.info("Convert into a name-sorted bam file, to get all reads with the same name adjacent to each other")
         pysam.sort("-o", basename + ".name-sorted.bam", "-n", self.input_alignment_file)
 
-        logging.info("Fixing sam file")
+        log.info("Fixing sam file")
         sam_file_discordant = pysam.AlignmentFile(basename + ".name-sorted.bam", "rb")
         header = sam_file_discordant.header
         header['RG'] = [
@@ -519,22 +519,22 @@ def convert(self, bam_file_discordant_fixed, temp_dir):
         self.reconstruct_alignments(alignments, sam_file_discordant, fh)
         fh.close()
 
-        logging.info("Converting fixed file into BAM")
+        log.info("Converting fixed file into BAM")
         fhq = open(basename + ".name-sorted.fixed.bam", "wb")
         fhq.write(pysam.view('-bS', basename + ".name-sorted.fixed.sam"))
         fhq.close()
 
-        logging.info("Sorting position based fixed file")
+        log.info("Sorting position based fixed file")
         pysam.sort("-o", basename + ".sorted.fixed.bam", basename + ".name-sorted.fixed.bam")
 
-        logging.info("Indexing the position sorted bam file")
+        log.info("Indexing the position sorted bam file")
         pysam.index(basename + ".sorted.fixed.bam")
 
-        logging.info("Cleaning up temp files")
+        log.info("Cleaning up temp files")
         for fname in [basename + ".name-sorted.bam", basename + ".name-sorted.fixed.sam", basename + ".name-sorted.fixed.bam"]:
-            logging.debug("=> " + fname)
+            log.debug("=> " + fname)
             os.remove(fname)
 
-        logging.info("Moving to final destination")
+        log.info("Moving to final destination")
         os.rename(basename + ".sorted.fixed.bam", bam_file_discordant_fixed)
         os.rename(basename + ".sorted.fixed.bam" + ".bai", bam_file_discordant_fixed + ".bai")

drdisco/IntronDecomposition.py

@@ -5,13 +5,13 @@
 
 from __init__ import MAX_ACCEPTABLE_INSERT_SIZE, MAX_ACCEPTABLE_ALIGNMENT_ERROR, MAX_GENOME_DISTANCE, MIN_SUBNET_ENTROPY, MIN_DISCO_PER_SUBNET_PER_NODE, MIN_SUPPORTING_READS_PER_SUBNET_PER_NODE
 
-import logging
 import math
 import operator
 
 import pysam
 import HTSeq
 
+from drdisco import log
 from .CigarAlignment import cigar_to_cigartuple
 
 from fuma.Fusion import STRAND_FORWARD, STRAND_REVERSE, STRAND_UNDETERMINED
@@ -601,7 +601,7 @@ def print_chain(self):  # pragma: no cover
         print "++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++"
 
     def generate_edge_idx(self):
-        logging.info("Creating edge index before pruning")
+        log.info("Creating edge index before pruning")
 
         edges = set()
         edges_tuple = []
@@ -624,7 +624,7 @@ def prune(self):
         """Does some 'clever' tricks to merge edges together and reduce data points
         """
         self.generate_edge_idx()
-        logging.info("Finding and merging other edges in close proximity (insert size)")
+        log.info("Finding and merging other edges in close proximity (insert size)")
 
         self.check_symmetry()
 
@@ -639,7 +639,7 @@ def prune(self):
 
             self.remove_edge(candidate)  # do not remove if splice junc exists?
 
-        logging.info("Pruned into " + str(len(candidates)) + " candidate edge(s)")
+        log.info("Pruned into " + str(len(candidates)) + " candidate edge(s)")
         return candidates
 
     def prune_edge(self, edge):
@@ -694,7 +694,7 @@ def rejoin_splice_juncs(self, splice_junctions):
 
         the goal is to add the splice juncs between the nodes
         """
-        logging.debug("Initiated")
+        log.debug("Initiated")
         k = 0
 
         def search(pos1):
@@ -736,7 +736,7 @@ def search(pos1):
                     splice_edges_had.add(splice_junction)
                     splice_edges_had.add(splice_junction.get_complement())
 
-        logging.info("Linked " + str(k) + " splice junction(s)")
+        log.info("Linked " + str(k) + " splice junction(s)")
 
     def extract_subnetworks_by_splice_junctions(self, thicker_edges, MIN_SCORE_FOR_EXTRACTING_SUBGRAPHS):
         """ Deze functie haalt recursief per edge een set van edges op die
@@ -751,7 +751,7 @@ def extract_subnetworks_by_splice_junctions(self, thicker_edges, MIN_SCORE_FOR_E
 en 1 met posities die groter zijn dan zichzelf. Hierdoor is een recursief
 terugloop probleem redelijk opgelost.
             """
-        logging.info("Initiated [MIN_SCORE_FOR_EXTRACTING_SUBGRAPHS=%i]" % MIN_SCORE_FOR_EXTRACTING_SUBGRAPHS)
+        log.info("Initiated [MIN_SCORE_FOR_EXTRACTING_SUBGRAPHS=%i]" % MIN_SCORE_FOR_EXTRACTING_SUBGRAPHS)
 
         thicker_edges.reverse()
         q = 0
@@ -793,7 +793,7 @@ def extract_subnetworks_by_splice_junctions(self, thicker_edges, MIN_SCORE_FOR_E
             subnetworks.append(SubGraph(q, subedges, left_splice_junctions, right_splice_junctions))
             self.remove_edge(start_point)
 
-        logging.info("Extracted %i subnetwork(s)" % len(subnetworks))
+        log.info("Extracted %i subnetwork(s)" % len(subnetworks))
         return subnetworks
 
 
@@ -1014,7 +1014,7 @@ def test_disco_alignment(alignment_file, require_fixed_bam_file):
             bam_fh.fetch()
         except:  # pragma: no cover
             fname = bam_fh.filename
-            logging.info('Indexing BAM file with pysam: ' + fname)  # create index if it does not exist
+            log.info('Indexing BAM file with pysam: ' + fname)  # create index if it does not exist
             bam_fh.close()
 
             pysam.index(fname)
@@ -1182,7 +1182,7 @@ def read_to_junction(read, rg, parsed_SA_tag, specific_type=None):
 
             return (None, None, None)
 
-        logging.debug("Parsing reads to obtain fusion gene and splice junctions")
+        log.debug("Parsing reads to obtain fusion gene and splice junctions")
         for read in self.pysam_fh.fetch():
             sa = self.parse_SA(read.get_tag('SA'))
             _chr = self.pysam_fh.get_reference_name(read.reference_id)
@@ -1237,7 +1237,7 @@ def read_to_junction(read, rg, parsed_SA_tag, specific_type=None):
                         else:
                             fusion_junctions.insert_edge(i_pos1, i_pos2, internal_edge[2], None)
 
-        logging.debug("alignment data loaded")
+        log.debug("alignment data loaded")
 
     def parse_pos(self, str_pos):
         _chr, _poss = str_pos.split(":", 2)
@@ -1450,7 +1450,7 @@ def merge_overlapping_subnets(self, subnets):
 
             merge all subnets in M into i, and remove the former subnets
         """
-        logging.info("initiated")
+        log.info("initiated")
 
         def sq_dist(vec):
             sum_of_squares = sum(pow(x, 2) for x in vec)
@@ -1579,11 +1579,11 @@ def tree_remove(genometree, subnet):
 
             new_subnets.append(subnet)
 
-        logging.info("Merged " + str(k) + " of the " + str(n) + " into " + str(len(new_subnets)) + " merged subnetwork(s)")
+        log.info("Merged " + str(k) + " of the " + str(n) + " into " + str(len(new_subnets)) + " merged subnetwork(s)")
         return new_subnets
 
     def filter_subnets(self, subnets):
-        logging.debug("init")
+        log.debug("init")
         k = 0
         for subnet in subnets:
             """Total of 8 reads is minimum, of which 2 must be
@@ -1618,5 +1618,5 @@ def filter_subnets(self, subnets):
             if len(subnet.discarded) > 0:
                 k += 1
 
-        logging.info("Filtered " + str(k) + " of the " + str(len(subnets)) + " subnetwork(s)")
+        log.info("Filtered " + str(k) + " of the " + str(len(subnets)) + " subnetwork(s)")
         return subnets

drdisco/__init__.py

@@ -28,13 +28,18 @@
     gmail dot com
 """
 
-__version_info__ = ('0', '3', '3')
+__version_info__ = ('0', '3', '4')
 __version__ = '.'.join(__version_info__) if (len(__version_info__) == 3) else '.'.join(__version_info__[0:3]) + "-" + __version_info__[3]
 __author__ = 'Youri Hoogstrate'
 __homepage__ = 'https://github.com/yhoogstrate/dr-disco'
 __license__ = 'GNU General Public License v3 (GPLv3)'
 __license_notice__ = 'License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>.\nThis is free software: you are free to change and redistribute it.\nThere is NO WARRANTY, to the extent permitted by law.'
+
+import logging
+import sys
 __log_format__ = "[%(filename)s:%(lineno)s - %(funcName)s()] %(asctime)s - %(levelname)s - %(message)s"
+logging.basicConfig(level=logging.DEBUG, format=__log_format__, stream=sys.stderr)  # bioconda seems to crash on stdout here..
+log = logging.getLogger(__name__)
 
 
 # parameters

tests/test_bam_extract.py

@@ -20,18 +20,13 @@
  along with this program. If not, see <http://www.gnu.org/licenses/>.
 """
 
-import drdisco
 from drdisco.IntronDecomposition import BAMExtract
 
 import unittest
-import logging
-import sys
 import filecmp
 import pysam
 import os
 
-logging.basicConfig(level=logging.DEBUG, format=drdisco.__log_format__, stream=sys.stdout)
-
 TEST_DIR = "tests/bam-extract/"
 T_TEST_DIR = "tmp/" + TEST_DIR
 

tests/test_cigaralignment.py

@@ -22,16 +22,11 @@
  along with this program. If not, see <http://www.gnu.org/licenses/>.
 """
 
-import drdisco
 from drdisco.CigarAlignment import CigarAlignment, cigar_to_cigartuple
 
 from fuma.Fusion import STRAND_REVERSE
 
 import unittest
-import logging
-import sys
-
-logging.basicConfig(level=logging.DEBUG, format=drdisco.__log_format__, stream=sys.stdout)
 
 
 class TestIntronicBreakDetection(unittest.TestCase):

tests/test_fix_chimeric_alignment.py

@@ -23,19 +23,14 @@
 """
 
 
-import drdisco
 from drdisco.ChimericAlignment import ChimericAlignment
 
 import unittest
-import logging
-import sys
 import subprocess
 import filecmp
 import pysam
 import os
 
-logging.basicConfig(level=logging.DEBUG, format=drdisco.__log_format__, stream=sys.stdout)
-
 subprocess.call(["bash", "tests/rm_bai_files.sh"])
 
 

tests/test_functional.py

@@ -20,16 +20,12 @@
  along with this program. If not, see <http://www.gnu.org/licenses/>.
 """
 
-import drdisco
 import unittest
-import logging
-import sys
 import filecmp
 import pysam
 import os
 import subprocess
 
-logging.basicConfig(level=logging.DEBUG, format=drdisco.__log_format__, stream=sys.stdout)
 # Nosetests doesn't use main()
 
 subprocess.call(["bash", "tests/rm_bai_files.sh"])

tests/test_intronic_break_detection.py

@@ -22,18 +22,13 @@
  along with this program. If not, see <http://www.gnu.org/licenses/>.
 """
 
-import drdisco
 from drdisco.IntronDecomposition import IntronDecomposition
 
 import unittest
-import logging
-import sys
 import subprocess
 import filecmp
 import os
 
-logging.basicConfig(level=logging.DEBUG, format=drdisco.__log_format__, stream=sys.stdout)
-
 subprocess.call(["bash", "tests/rm_bai_files.sh"])
 
 TEST_DIR = "tests/detect-intronic/"

tests/test_logo_sequence.py

@@ -20,16 +20,11 @@
  along with this program. If not, see <http://www.gnu.org/licenses/>.
 """
 
-import drdisco
-import logging
-import sys
 import os
 import unittest
 import subprocess
 import filecmp
 
-logging.basicConfig(level=logging.DEBUG, format=drdisco.__log_format__, stream=sys.stdout)
-
 subprocess.call(["bash", "tests/rm_bai_files.sh"])
 
 TEST_DIR = "tests/logo-sequence/"