Merge pull request #14 from galaxy-genome-annotation/fix-biopython

Author: hexylena

chado/export/__init__.py

@@ -12,7 +12,6 @@
 from BCBio import GFF
 
 from Bio import SeqIO
-from Bio.Alphabet import IUPAC
 from Bio.Seq import Seq
 from Bio.SeqFeature import FeatureLocation as BioFeatureLocation
 from Bio.SeqFeature import SeqFeature
@@ -85,7 +84,7 @@ def export_gff3(self, organism_id):
 
         for org in res:
             # TODO: can we do this properly?
-            seq = Seq("A" * 1, IUPAC.unambiguous_dna)
+            seq = Seq("A" * 1)
 
             # Annotation features
             features = self.session.query(self.model.feature, self.model.featureloc) \
@@ -230,7 +229,7 @@ def export_gbk(self, organism_id):
 
         sys.stderr.write("Processing %s sequences\n" % res.count())
         for org in res:
-            seq = Seq('', IUPAC.unambiguous_dna)
+            seq = Seq('')
 
             record_features = []
             features = self.ci.session.query(self.model.feature, self.model.featureloc) \
@@ -245,7 +244,7 @@ def export_gbk(self, organism_id):
                 # Sequence containing feature
                 if feature.residues:
                     # This seems bad? What if multiple things have seqs?
-                    seq = Seq(feature.residues, IUPAC.unambiguous_dna)
+                    seq = Seq(feature.residues)
                 else:
                     qualifiers = {
                         self.ci.get_cvterm_name(prop.type_id): prop.value for prop in
@@ -266,6 +265,7 @@ def export_gbk(self, organism_id):
                 seq, id=org.common_name,
                 name=org.common_name,
                 description="%s %s" % (org.genus, org.species),
+                annotations={"molecule_type": "DNA"},
             )
             record.features = record_features
 

requirements.txt

@@ -1,6 +1,6 @@
 sqlalchemy
 psycopg2-binary
-biopython
+biopython>=1.78
 bcbio-gff
 wrapt
 click