Ignore inapplicable DeprecationWarning

Tests/test_orthofinder.py

@@ -177,12 +177,14 @@ def setUpClass(cls):
 
 
 #    @unittest.skipIf(__skipLongTests__, "Only performing quick tests")     
-    def test_fromfasta(self):
+    def test_fromfasta_full(self):
         currentResultsDir = exampleFastaDir + "Results_%s/" % datetime.date.today().strftime("%b%d") 
         expectedCSVFile = currentResultsDir + "Orthogroups.csv"
         with CleanUp([], [], [currentResultsDir, ]):
-            self.stdout, self.stderr = self.RunOrthoFinder("-f %s -g" % exampleFastaDir)
+            self.stdout, self.stderr = self.RunOrthoFinder("-f %s" % exampleFastaDir)
             self.CheckStandardRun(self.stdout, self.stderr, goldResultsDir_smallExample, expectedCSVFile)  
+            expectedTreeFN = exampleFastaDir + ("Results_%s/" % datetime.date.today().strftime("%b%d")) + ("Orthologues_%s/" % datetime.date.today().strftime("%b%d")) + "Gene_Trees/OG0000000_tree.txt"
+            self.assertTrue(os.path.exists(expectedTreeFN))
         self.test_passed = True         
 
     def test_fromfasta_threads(self):

orthofinder/orthofinder.py

@@ -475,25 +475,29 @@ def ConnectCognates(seqsInfo, fileInfo, iSpecies):
 
     @staticmethod 
     def Worker_ConnectCognates(cmd_queue):
-        while True:
-            try:
-                args = cmd_queue.get(True, 1)
-                WaterfallMethod.ConnectCognates(*args)
-            except Queue.Empty:
-                return  
+        with warnings.catch_warnings():         
+            warnings.simplefilter("ignore")
+            while True:
+                try:
+                    args = cmd_queue.get(True, 1)
+                    WaterfallMethod.ConnectCognates(*args)
+                except Queue.Empty:
+                    return  
 
     @staticmethod
     def WriteGraphParallel(seqsInfo, fileInfo):
-        with open(fileInfo.graphFilename + "_header", 'wb') as graphFile:
-            graphFile.write("(mclheader\nmcltype matrix\ndimensions %dx%d\n)\n" % (seqsInfo.nSeqs, seqsInfo.nSeqs)) 
-            graphFile.write("\n(mclmatrix\nbegin\n\n") 
-        pool = mp.Pool()
-        pool.map(WriteGraph_perSpecies, [(seqsInfo, fileInfo, iSpec) for iSpec in xrange(seqsInfo.nSpecies)])
-        subprocess.call("cat " + fileInfo.graphFilename + "_header " + " ".join([fileInfo.graphFilename + "_%d" % iSp for iSp in xrange(seqsInfo.nSpecies)]) + " > " + fileInfo.graphFilename, shell=True)
-        # Cleanup
-        os.remove(fileInfo.graphFilename + "_header")
-        for iSp in xrange(seqsInfo.nSpecies): os.remove(fileInfo.graphFilename + "_%d" % iSp)
-        DeleteMatrices(fileInfo) 
+        with warnings.catch_warnings():         
+            warnings.simplefilter("ignore")
+            with open(fileInfo.graphFilename + "_header", 'wb') as graphFile:
+                graphFile.write("(mclheader\nmcltype matrix\ndimensions %dx%d\n)\n" % (seqsInfo.nSeqs, seqsInfo.nSeqs)) 
+                graphFile.write("\n(mclmatrix\nbegin\n\n") 
+            pool = mp.Pool()
+            pool.map(WriteGraph_perSpecies, [(seqsInfo, fileInfo, iSpec) for iSpec in xrange(seqsInfo.nSpecies)])
+            subprocess.call("cat " + fileInfo.graphFilename + "_header " + " ".join([fileInfo.graphFilename + "_%d" % iSp for iSp in xrange(seqsInfo.nSpecies)]) + " > " + fileInfo.graphFilename, shell=True)
+            # Cleanup
+            os.remove(fileInfo.graphFilename + "_header")
+            for iSp in xrange(seqsInfo.nSpecies): os.remove(fileInfo.graphFilename + "_%d" % iSp)
+            DeleteMatrices(fileInfo) 
 
     @staticmethod
     def GetMostDistant_s(RBH, B, seqsInfo, iSpec):

orthofinder/scripts/get_orthologues.py

@@ -36,6 +36,7 @@
 import itertools
 import multiprocessing as mp
 import Queue
+import warnings
 
 import util
 import tree
@@ -107,8 +108,8 @@ def SequenceDict(self):
                     util.Fail()
                 else:
                     print("Tried to use only the first part of the accession in order to list the sequences in each orthogroup\nmore concisely but these were not unique. The full accession line will be used instead.\n")     
-                    self.seqIDsEx = util.FullAccession(self.seqIDsFN).GetIDToNameDict()
-        return self.seqIDsEx
+                    self.seqIDsEx = util.FullAccession(self.seqIDsFN)
+        return self.seqIDsEx.GetIDToNameDict()
 
     def SpeciesDict(self):
         d = self.speciesIDsEx.GetIDToNameDict()
@@ -245,18 +246,20 @@ def __init__(self, ogSet, outD, nProcesses):
         # Check files exist
 
     def ReadAndPickle(self): 
-        cmd_queue = mp.Queue()
-        i = 0
-        for iSp in xrange(len(self.ogSet.seqsInfo.speciesToUse)):
-            for jSp in xrange(len(self.ogSet.seqsInfo.speciesToUse)):
-                cmd_queue.put((i, (iSp, jSp)))           
-                i+=1
-        runningProcesses = [mp.Process(target=Worker_BlastScores, args=(cmd_queue, self.ogSet.seqsInfo, self.ogSet.fileInfo, nThreads, i)) for i_ in xrange(nThreads)]
-        for proc in runningProcesses:
-            proc.start()
-        for proc in runningProcesses:
-            while proc.is_alive():
-                proc.join() 
+        with warnings.catch_warnings():         
+            warnings.simplefilter("ignore")
+            cmd_queue = mp.Queue()
+            i = 0
+            for iSp in xrange(len(self.ogSet.seqsInfo.speciesToUse)):
+                for jSp in xrange(len(self.ogSet.seqsInfo.speciesToUse)):
+                    cmd_queue.put((i, (iSp, jSp)))           
+                    i+=1
+            runningProcesses = [mp.Process(target=Worker_BlastScores, args=(cmd_queue, self.ogSet.seqsInfo, self.ogSet.fileInfo, nThreads, i)) for i_ in xrange(nThreads)]
+            for proc in runningProcesses:
+                proc.start()
+            for proc in runningProcesses:
+                while proc.is_alive():
+                    proc.join() 
 
     def NumberOfSequences(self, species):
         ids = self.ogSet.SequenceDict()
@@ -269,25 +272,27 @@ def GetOGMatrices(self):
         ogMatrices contains matrix M for each OG where:
             Mij = 0.5*max(Bij, Bmin_i)/Bmax_i
         """
-        ogs = self.ogSet.OGs()
-        ogsPerSpecies = [[[(g, i) for i, g in enumerate(og) if g.iSp == iSp] for iSp in self.species] for og in ogs]
-        nGenes = [len(og) for og in ogs]
-        nSeqs = self.NumberOfSequences(self.species)
-        ogMatrices = [np.zeros((n, n)) for n in nGenes]
-        for iiSp, sp1 in enumerate(self.species):
-            util.PrintTime("Processing species %d" % sp1)
-            Bs = [matrices.LoadMatrix("Bit", self.ogSet.fileInfo, iiSp, jjSp) for jjSp in xrange(len(self.species))]
-            mins = np.ones((nSeqs[sp1], 1), dtype=np.float64)*9e99 
-            maxes = np.zeros((nSeqs[sp1], 1), dtype=np.float64)
-            for B, sp2 in zip(Bs, self.species):
-                mins = np.minimum(mins, lil_min(B))
-                maxes = np.maximum(maxes, lil_max(B))
-            for jjSp, B  in enumerate(Bs):
-                for og, m in zip(ogsPerSpecies, ogMatrices):
-                    for gi, i in og[iiSp]:
-                        for gj, j in og[jjSp]:
-                                m[i, j] = 0.5*max(B[gi.iSeq, gj.iSeq], mins[gi.iSeq]) /  maxes[gi.iSeq]
-        return ogs, ogMatrices
+        with warnings.catch_warnings():         
+            warnings.simplefilter("ignore")
+            ogs = self.ogSet.OGs()
+            ogsPerSpecies = [[[(g, i) for i, g in enumerate(og) if g.iSp == iSp] for iSp in self.species] for og in ogs]
+            nGenes = [len(og) for og in ogs]
+            nSeqs = self.NumberOfSequences(self.species)
+            ogMatrices = [np.zeros((n, n)) for n in nGenes]
+            for iiSp, sp1 in enumerate(self.species):
+                util.PrintTime("Processing species %d" % sp1)
+                Bs = [matrices.LoadMatrix("Bit", self.ogSet.fileInfo, iiSp, jjSp) for jjSp in xrange(len(self.species))]
+                mins = np.ones((nSeqs[sp1], 1), dtype=np.float64)*9e99 
+                maxes = np.zeros((nSeqs[sp1], 1), dtype=np.float64)
+                for B, sp2 in zip(Bs, self.species):
+                    mins = np.minimum(mins, lil_min(B))
+                    maxes = np.maximum(maxes, lil_max(B))
+                for jjSp, B  in enumerate(Bs):
+                    for og, m in zip(ogsPerSpecies, ogMatrices):
+                        for gi, i in og[iiSp]:
+                            for gj, j in og[jjSp]:
+                                    m[i, j] = 0.5*max(B[gi.iSeq, gj.iSeq], mins[gi.iSeq]) /  maxes[gi.iSeq]
+            return ogs, ogMatrices
 
     def DeleteBlastMatrices(self):
         for f in glob.glob(self.ogSet.fileInfo.outputDir + "Bit*_*.pic"):

orthofinder/scripts/orthologues_from_recon_trees.py

@@ -96,7 +96,7 @@ def one_to_one_efficient(orthodict, genenumbers, speciesLabels, iSpecies, output
             jSpLabel,jGene = map(int,Gene.split('_'))
             jSp = speciesLabelsReverse[jSpLabel]
             if iSpecies > jSp:
-                matrixlist[jSp][iGene, jGene]  = 1
+                matrixlist[jSp][iGene, jGene] = 1
     for j, m in enumerate(matrixlist):    
         with open(outputDir + 'ortholog_%d_%d_matrix.pic' % (iSpecies, j), 'wb') as file:
             pic.dump(m, file)
@@ -128,7 +128,7 @@ def WriteOrthologues(resultsDir, spec1, spec2, orthologues, ogSet):
 def GetOrthologues(orig_matrix, orig_matrix_csc, index):
     orthologues = orig_matrix.getrowview(index).nonzero()[1]
     index = orthologues[0]
-    originalSpeciesGenes = orig_matrix_csc.getcol(index).nonzero()[0]
+    originalSpeciesGenes = orig_matrix_csc.getcol(index).nonzero()[0]  # SparseEfficiencyWarning: changing the sparsity structure of a csc_matrix is expensive. lil_matrix is more efficient
     return (originalSpeciesGenes, orthologues)
 
 #Takes in the output of multiply, finds all of the orthology relationships which it writes to textfiles and returns the number of each type of orthology relationship.