Major bug fixed

The contigs identifier of some assemblers like unicycler is just a number, I opted to standardize all searches using strings
EnzoAndree · Feb 15, 2021 · cb404fd · cb404fd
1 parent 8dc6f45
commit cb404fd
Show file tree

Hide file tree

Showing 3 changed files with 15 additions and 12 deletions.
diff --git a/bin/fastmlst b/bin/fastmlst
@@ -48,7 +48,7 @@ def runMLST(margument):
 
 
 if __name__ == '__main__':
-    V = '%(prog)s v0.0.11'
+    V = '%(prog)s v0.0.12'
     parser = argparse.ArgumentParser()
     parser.add_argument(type=str, nargs='*', dest='genomes')
     parser.add_argument('-t', '--threads', type=int, default=cpu_count(),
@@ -64,9 +64,9 @@ if __name__ == '__main__':
                         help='Set a scheme target (I am not dumb, let me choose a scheme by myself!)')
     parser.add_argument('--scheme-list', action='store_true',
                         help='Show all schemes supported')
-    parser.add_argument('-fo', '--fastaoutput', type=str, default='mlst.fasta',
+    parser.add_argument('-fo', '--fastaoutput', type=str, default='',
                         help='File name of the concatenated alleles output' +
-                        ' (default mlst.fasta)')
+                        ' (default "")')
     parser.add_argument('-to', '--tableoutput', type=str, default=stdout,
                         help='File name of the MLST table output' +
                         ' (default STDOUT)')
@@ -201,14 +201,15 @@ if __name__ == '__main__':
             str_alleles += genome.str_st
             str_alleles += '\n'
             dict_alleles.append(genome.dict_st)
-    if args.fasta2line:
-        SeqIO.write(fastaconcat, args.fastaoutput, 'fasta-2line')
-        if args.novel:
-            SeqIO.write(fastanovelconcat, args.novel, 'fasta-2line')
-    else:
-        SeqIO.write(fastaconcat, args.fastaoutput, 'fasta')
-        if args.novel:
-            SeqIO.write(fastanovelconcat, args.novel, 'fasta')
+    # FastMLSTv0.0.12 by default do not write the fasta file 
+        if args.fasta2line and args.fastaoutput != '':
+            SeqIO.write(fastaconcat, args.fastaoutput, 'fasta-2line')
+            if args.novel:
+                SeqIO.write(fastanovelconcat, args.novel, 'fasta-2line')
+        elif args.fastaoutput != '':
+            SeqIO.write(fastaconcat, args.fastaoutput, 'fasta')
+            if args.novel:
+                SeqIO.write(fastanovelconcat, args.novel, 'fasta')
 
     # output formated
     if type(args.tableoutput) == str:

diff --git a/fastmlst/mlst.py b/fastmlst/mlst.py
@@ -146,6 +146,8 @@ def blast_filter(self, blast_out):
         dfblast['coverage'] = (dfblast.length - dfblast.gaps) / dfblast.slen
         dfblast['identity'] = (dfblast.nident - dfblast.gaps) / dfblast.slen # this is a 'global' %identity
         dfblast[toint] = dfblast[toint].astype(int)
+        # Bug Fixed on FastMLSTv0.0.12: The contigs identifier of some assemblers like unicycler is just a number, I opted to standardize all searches using strings
+        dfblast['qseqid'] = dfblast['qseqid'].astype(str)
         # don't even look at the hits below these cov's and identities
         dfblast = dfblast.loc[(dfblast['coverage'] <= 1) & (dfblast['coverage'] >= self.coverage) & (dfblast['identity'] >= self.identity)] # insertions can not be processed properly yet
         # dfblast = dfblast.loc[dfblast['slen'] >= dfblast['length']] # if have an insertion slen < length

diff --git a/setup.py b/setup.py
@@ -1,7 +1,7 @@
 from setuptools import setup
 
 setup(name='fastmlst',
-    version='0.0.11',
+    version='0.0.12',
     description='A multi-core tool for multilocus sequence typing of draft genome assemblies using PubMLST typing schemes',
     url='https://github.com/EnzoAndree/FastMLST',
     author='Enzo Guerrero-Araya',