For #280: fixed the previous error.

Mikado/serializers/blast_serializer/utils.py

@@ -6,7 +6,7 @@
 import numpy as np
 
 valid_letters = 'ACDEFGHIKLMNPQRSTVWYBXZJUO'
-letters = np.array(list(valid_letters))
+letters = np.array(list(valid_letters), dtype=np.str_)
 
 
 __author__ = 'Luca Venturini'
@@ -35,7 +35,7 @@ def prepare_hsp(hsp, counter, qmultiplier=1, tmultiplier=1):
 
     hsp_dict = dict()
     # We must start from 1, otherwise MySQL crashes as its indices start from 1 not 0
-    match, identical_positions, positives = _prepare_aln_strings(hsp, qmultiplier=qmultiplier, tmultiplier=tmultiplier)
+    match, identical_positions, positives = _prepare_aln_strings(hsp, qmultiplier=qmultiplier)
     hsp_dict["counter"] = counter + 1
     hsp_dict["query_hsp_start"] = hsp.query_start
     hsp_dict["query_hsp_end"] = hsp.query_end
@@ -56,7 +56,7 @@ def _np_grouper(data):
     return np.array(np.split(data, np.where(np.diff(data) != 1)[0] + 1))
 
 
-def _prepare_aln_strings(hsp, qmultiplier=1, tmultiplier=1):
+def _prepare_aln_strings(hsp, qmultiplier=1):
 
     """This private method calculates the identical positions, the positives, and a re-factored match line
     starting from the HSP.
@@ -66,7 +66,7 @@ def _prepare_aln_strings(hsp, qmultiplier=1, tmultiplier=1):
     lett_array = np.array([
         list(str(hsp.query.seq)),
         list(hsp.aln_annotation["similarity"]),
-        list(str(hsp.hit.seq))])
+        list(str(hsp.hit.seq))], dtype=np.str_)
 
     match = lett_array[1]
     match[np.where(~((lett_array[1] == "+") | (np.isin(lett_array[1], letters))))] = " "

Mikado/serializers/blast_serializer/xml_serialiser.py

@@ -88,8 +88,7 @@ def xml_pickler(json_conf, filename, default_header,
                 qmult, tmult = None, None
                 for query_counter, record in enumerate(opened, start=1):
                     if qmult is None:
-                        qmult, tmult = XmlSerializer._get_multipliers(record)
-
+                        qmult, tmult = XmlSerializer.get_multipliers(record)
                     hits, hsps, cache = objectify_record(
                         session, record, [], [], cache, max_target_seqs=max_target_seqs,
                         qmult=qmult, tmult=tmult)
@@ -547,15 +546,13 @@ def get_multipliers(record):
         Private quick method to determine the multipliers for a BLAST alignment
         according to the application present in the record.
         :param record:
-        :type record: Bio.Blast.Record.Blast
+        :type record: Bio.SearchIO.Record
         :return:
         """
 
         q_mult, h_mult = 1, 1
 
-        # application = record.application.upper()
-        application = record.application.upper()
-
+        application = record.program.upper()
         if application in ("BLASTN", "TBLASTX", "BLASTP"):
             q_mult = 1
             h_mult = 1

Mikado/tests/test_system_calls.py

@@ -1677,6 +1677,53 @@ def setUp(self):
     def setUpClass(cls):
         cls.fai = pysam.FastaFile(pkg_resources.resource_filename("Mikado.tests", "chr5.fas.gz"))
 
+    def test_subprocess_single(self):
+
+        xml = pkg_resources.resource_filename("Mikado.tests", "chunk-001-proteins.xml.gz")
+        transcripts = pkg_resources.resource_filename("Mikado.tests", "mikado_prepared.fasta")
+        junctions = pkg_resources.resource_filename("Mikado.tests", "junctions.bed")
+        orfs = pkg_resources.resource_filename("Mikado.tests", "transcripts.fasta.prodigal.gff3")
+        uniprot = pkg_resources.resource_filename("Mikado.tests", "uniprot_sprot_plants.fasta.gz")
+        mobjects = 300  # Let's test properly the serialisation for BLAST
+
+        dir = tempfile.TemporaryDirectory()
+        json_file = os.path.join(dir.name, "mikado.yaml")
+        db = os.path.join(dir.name, "mikado.db")
+        log = os.path.join(dir.name, "serialise.log")
+        uni_out = os.path.join(dir.name, "uniprot_sprot_plants.fasta")
+        with gzip.open(uniprot, "rb") as uni, open(uni_out, "wb") as uni_out_handle:
+            uni_out_handle.write(uni.read())
+
+        with open(json_file, "wt") as json_handle:
+            sub_configure.print_config(yaml.dump(self.json_conf, default_flow_style=False),
+                                                      json_handle)
+        # Set up the command arguments
+        for procs in (1,):
+            with self.subTest(proc=procs):
+                sys.argv = [str(_) for _ in ["mikado", "serialise", "--json-conf", json_file,
+                            "--transcripts", transcripts, "--blast_targets", uni_out,
+                            "--orfs", orfs, "--junctions", junctions, "--xml", xml,
+                            "-p", procs, "-mo", mobjects, db, "--log", log, "--seed", "1078"]]
+                pkg_resources.load_entry_point("Mikado", "console_scripts", "mikado")()
+                logged = [_.rstrip() for _ in open(log)]
+
+                self.assertTrue(os.path.exists(db))
+                conn = sqlite3.connect(db)
+                cursor = conn.cursor()
+                self.assertEqual(cursor.execute("select count(*) from hit").fetchall()[0][0], 562, logged)
+                self.assertEqual(cursor.execute("select count(*) from hsp").fetchall()[0][0], 669)
+                self.assertEqual(cursor.execute("select count(distinct(query_id)) from hsp").fetchall()[0][0], 71)
+                self.assertEqual(cursor.execute("select count(distinct(query_id)) from hit").fetchall()[0][0], 71)
+                self.assertEqual(cursor.execute("select count(distinct(target_id)) from hsp").fetchall()[0][0], 32)
+                self.assertEqual(cursor.execute("select count(distinct(target_id)) from hit").fetchall()[0][0], 32)
+                self.assertEqual(cursor.execute("select count(*) from junctions").fetchall()[0][0], 372)
+                self.assertEqual(cursor.execute("select count(distinct(chrom_id)) from junctions").fetchall()[0][0], 2)
+                self.assertEqual(cursor.execute("select count(*) from orf").fetchall()[0][0], 169,
+                                 "\n".join(logged))
+                self.assertEqual(cursor.execute("select count(distinct(query_id)) from orf").fetchall()[0][0], 81)
+                os.remove(db)
+        dir.cleanup()
+
     def test_subprocess_multi(self):
 
         xml = pkg_resources.resource_filename("Mikado.tests", "chunk-001-proteins.xml.gz")
@@ -1698,7 +1745,7 @@ def test_subprocess_multi(self):
             sub_configure.print_config(yaml.dump(self.json_conf, default_flow_style=False),
                                                       json_handle)
         # Set up the command arguments
-        for procs in (1, 3):
+        for procs in (3,):
             with self.subTest(proc=procs):
                 sys.argv = [str(_) for _ in ["mikado", "serialise", "--json-conf", json_file,
                             "--transcripts", transcripts, "--blast_targets", uni_out,