fix for EVM path issue and better logging for some modules

Author: Jon Palmer

bin/funannotate-functional.py

@@ -217,8 +217,8 @@ def runIPRpython(Input):
         Transcripts = os.path.join(outputdir, 'annotate_misc', 'genome.transcripts.fasta')
         lib.gb2output(genbank, Proteins, Transcripts, Scaffolds)
 
-#get absolute path for all input so there are no problems later
-Scaffolds, Proteins, Transcripts, GFF = [os.path.abspath(i) for i in [Scaffolds, Proteins, Transcripts, GFF]] #suggestion via GitHub
+#get absolute path for all input so there are no problems later, not using Transcripts yet could be error? so take out here
+Scaffolds, Proteins, GFF = [os.path.abspath(i) for i in [Scaffolds, Proteins, GFF]] #suggestion via GitHub
 '''
 for i in Scaffolds, Proteins, Transcripts, GFF:
     i = os.path.abspath(i)

bin/funannotate-p2g.py

@@ -25,10 +25,13 @@ def tblastnFilter(input, query, cpus, output):
     if not os.path.exists(output):
         os.makedirs(output)
     #start by formatting blast db/dustmasker filtered format
-    subprocess.call(['dustmasker', '-in', input, '-infmt', 'fasta', '-parse_seqids', '-outfmt', 'maskinfo_asn1_bin', '-out', 'genome_dust.asnb'], cwd = output, stdout = FNULL, stderr = FNULL)
-    subprocess.call(['makeblastdb', '-in', input, '-dbtype', 'nucl', '-parse_seqids', '-mask_data', 'genome_dust.asnb', '-out', 'genome'], cwd = output, stdout = FNULL, stderr = FNULL)
-    #okay, now run tblastn using uniprot proteins
-    subprocess.call(['tblastn', '-num_threads', str(cpus), '-db', 'genome', '-query', query, '-max_target_seqs', '1', '-db_soft_mask', '11', '-threshold', '999', '-max_intron_length', MaxIntron, '-evalue', '1e-10', '-outfmt', '6', '-out', 'filter.tblastn.tab'], cwd = output, stdout = FNULL, stderr = FNULL)
+    cmd = ['dustmasker', '-in', input, '-infmt', 'fasta', '-parse_seqids', '-outfmt', 'maskinfo_asn1_bin', '-out', 'genome_dust.asnb']
+    lib.runSubprocess(cmd, output, lib.log)
+    cmd = ['makeblastdb', '-in', input, '-dbtype', 'nucl', '-parse_seqids', '-mask_data', 'genome_dust.asnb', '-out', 'genome']
+    lib.runSubprocess(cmd, output, lib.log)
+    cmd = ['tblastn', '-num_threads', str(cpus), '-db', 'genome', '-query', query, '-max_target_seqs', '1', '-db_soft_mask', '11', '-threshold', '999', '-max_intron_length', MaxIntron, '-evalue', '1e-10', '-outfmt', '6', '-out', 'filter.tblastn.tab']
+    lib.runSubprocess(cmd, output, lib.log)
+    
     #now parse through results, generating a list for exonerate function
     with open(os.path.join(output, 'filter.tblastn.tab')) as input:
         reader = csv.reader(input, delimiter='\t')

bin/funannotate-predict.py

@@ -887,10 +887,14 @@ def __init__(self, prog):
 
 #parse entire EVM command to script
 if Exonerate and Transcripts:
+    Transcripts = os.path.abspath(Transcripts)
+    Exonerate = os.path.abspath(Exonerate)
     evm_cmd = [sys.executable, EVM_script, os.path.join(args.out, 'logfiles', 'funannotate-EVM.log'), str(args.cpus), '--genome', MaskGenome, '--gene_predictions', Predictions, '--protein_alignments', Exonerate, '--transcript_alignments', Transcripts, '--weights', Weights, '--min_intron_length', str(args.min_intronlen), EVM_out]
 elif not Exonerate and Transcripts:
+    Transcripts = os.path.abspath(Transcripts)
     evm_cmd = [sys.executable, EVM_script, os.path.join(args.out, 'logfiles', 'funannotate-EVM.log'),str(args.cpus), '--genome', MaskGenome, '--gene_predictions', Predictions, '--transcript_alignments', Transcripts, '--weights', Weights, '--min_intron_length', str(args.min_intronlen), EVM_out]
 elif not Transcripts and Exonerate:
+    Exonerate = os.path.abspath(Exonerate)
     evm_cmd = [sys.executable, EVM_script, os.path.join(args.out, 'logfiles', 'funannotate-EVM.log'), str(args.cpus), '--genome', MaskGenome, '--gene_predictions', Predictions, '--protein_alignments', Exonerate, '--weights', Weights, '--min_intron_length', str(args.min_intronlen), EVM_out]
 elif not any([Transcripts,Exonerate]):
     evm_cmd = [sys.executable, EVM_script, os.path.join(args.out, 'logfiles', 'funannotate-EVM.log'), str(args.cpus), '--genome', MaskGenome, '--gene_predictions', Predictions, '--weights', Weights, '--min_intron_length', str(args.min_intronlen), EVM_out]

bin/funannotate-runEVM.py

@@ -76,7 +76,8 @@ def safe_run(*args, **kwargs):
 
 #split partitions
 lib.log.info("Setting up EVM partitions")
-subprocess.call(cmd1, cwd = tmpdir, stdout = FNULL, stderr = FNULL)
+lib.runSubprocess(cmd1, tmpdir, lib.log)
+#subprocess.call(cmd1, cwd = tmpdir, stdout = FNULL, stderr = FNULL)
 #check output
 lib.checkinputs(os.path.join(tmpdir, 'partitions_list.out'))
 
@@ -118,11 +119,14 @@ def safe_run(*args, **kwargs):
 
 #now combine the paritions
 lib.log.info("Combining partitioned EVM outputs")
-subprocess.call([perl, Combine, '--partitions', 'partitions_list.out', '--output_file_name', 'evm.out'], cwd = tmpdir, stdout = FNULL, stderr = FNULL)
+partitioncmd = [perl, Combine, '--partitions', 'partitions_list.out', '--output_file_name', 'evm.out']
+lib.runSubprocess(partitioncmd, tmpdir, lib.log)
+#subprocess.call([perl, Combine, '--partitions', 'partitions_list.out', '--output_file_name', 'evm.out'], cwd = tmpdir, stdout = FNULL, stderr = FNULL)
 
 #now convert to GFF3
 lib.log.info("Converting EVM output to GFF3")
-subprocess.call(cmd5, cwd = tmpdir, stdout = FNULL, stderr = FNULL)
+lib.runSubprocess(cmd5, tmpdir, lib.log)
+#subprocess.call(cmd5, cwd = tmpdir, stdout = FNULL, stderr = FNULL)
 
 #now concatenate all GFF3 files together for a genome then
 lib.log.info("Collecting all EVM results")

funannotate.py

@@ -31,7 +31,7 @@ def fmtcols(mylist, cols):
              for i in range(0,num_lines))
     return "\n".join(lines)
 
-version = '0.3.11'
+version = '0.3.12'
 
 default_help = """
 Usage:       funannotate <command> <arguments>

lib/library.py

@@ -55,7 +55,16 @@ class colr:
     GRN = '\033[92m'
     END = '\033[0m'
     WARN = '\033[93m'
-    
+
+def runSubprocess(cmd, dir, logfile):
+    logfile.debug(' '.join(cmd))
+    proc = subprocess.Popen(cmd, cwd=dir, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+    stdout, stderr = proc.communicate()
+    if stdout:
+        logfile.debug(stdout)
+    if stderr:
+        logfile.debug(stderr)
+
 def hashfile(afile, hasher, blocksize=65536):
     buf = afile.read(blocksize)
     while len(buf) > 0: