Avoid blank reads, as part of #442.

They caused a TypeError in iva assembly.
cfe-lab · Oct 29, 2018 · 41d7a18 · 41d7a18
1 parent e1903d6
commit 41d7a18
Show file tree

Hide file tree

Showing 4 changed files with 22 additions and 9 deletions.
diff --git a/micall/core/denovo.py b/micall/core/denovo.py
@@ -7,7 +7,7 @@
 from glob import glob
 from io import StringIO
 from shutil import rmtree
-from subprocess import Popen, call
+from subprocess import Popen
 from tempfile import mkdtemp
 
 from Bio import SeqIO
@@ -142,6 +142,7 @@ def denovo(fastq1_path, fastq2_path, contigs, work_dir='.', merged_contigs_csv=N
 
 
 if __name__ == '__main__':
+    logging.basicConfig(level=logging.INFO)
     parser = argparse.ArgumentParser()
     parser.add_argument('fastq1')
     parser.add_argument('fastq2')

diff --git a/micall/core/trim_fastqs.py b/micall/core/trim_fastqs.py
@@ -86,6 +86,7 @@ def trim(original_fastq_filenames,
                      '-o', trimmed_fastq_filenames[0],
                      '-p', trimmed_fastq_filenames[1],
                      '--quiet',
+                     '--minimum-length', '1',
                      censored_filenames[0],
                      censored_filenames[1]]
     cut_adapt.check_output(cutadapt_args)

diff --git a/micall/tests/microtest/make_sample.py b/micall/tests/microtest/make_sample.py
@@ -65,7 +65,7 @@ def main():
                 ref_nuc_section = ''.join(ref_nuc_section)
                 if fastq_file.is_reversed:
                     ref_nuc_section = reverse_and_complement(ref_nuc_section)
-                phred_scores = 'A' * (ref_end-ref_start)
+                phred_scores = 'A' * len(ref_nuc_section)
                 file_num = '2' if fastq_file.is_reversed else '1'
                 for cluster in range(section.count):
                     f.write('@M01234:01:000000000-AAAAA:1:1101:{}:{:04} {}:N:0:1\n'.format(

diff --git a/micall/utils/denovo_simplify.py b/micall/utils/denovo_simplify.py
@@ -34,7 +34,7 @@ def parse_args():
 
 
 class MicallDD(DD):
-    test_names = ('one_contig', 'multiple_genotypes')
+    test_names = ('one_contig', 'multiple_genotypes', 'type_error')
 
     def __init__(self,
                  filename1,
@@ -65,7 +65,6 @@ def __init__(self,
 
     def _test(self, read_indexes):
         read_count = len(read_indexes)
-        read_indexes = reversed(read_indexes)
         self.write_simple_fastq(self.simple1, self.simple2, read_indexes)
         workdir = os.path.dirname(self.simple1)
         os.chdir(workdir)
@@ -77,30 +76,35 @@ def _test(self, read_indexes):
                  self.bad_cycles_filename,
                  (trimmed_filename1, trimmed_filename2),
                  use_gzip=False)
+            exception = None
             # noinspection PyBroadException
             try:
                 denovo(trimmed_filename1, trimmed_filename2, contigs_csv, workdir)
-            except Exception:
+            except Exception as ex:
                 logger.warning('Assembly failed.', exc_info=True)
-                return DD.UNRESOLVED
+                exception = ex
             contigs_csv.seek(0)
 
-            result = self.get_result(contigs_csv, read_count)
+            result = self.get_result(contigs_csv, read_count, exception)
             if result == DD.FAIL:
                 os.rename(self.simple1, self.best1)
                 os.rename(self.simple2, self.best2)
             return result
 
     @staticmethod
-    def check_one_contig(contigs_csv, read_count):
+    def check_one_contig(contigs_csv, read_count, exception):
+        if exception is not None:
+            return DD.UNRESOLVED
         contig_count = len(contigs_csv.readlines())
         logger.debug('Result: %d contigs from %d reads.',
                      contig_count,
                      read_count)
         return DD.FAIL if contig_count == 1 else DD.PASS
 
     @staticmethod
-    def check_multiple_genotypes(contigs_csv, read_count):
+    def check_multiple_genotypes(contigs_csv, read_count, exception):
+        if exception is not None:
+            return DD.UNRESOLVED
         reader = DictReader(contigs_csv)
         genotypes = sorted({row['genotype'] for row in reader})
         genotype_count = len(genotypes)
@@ -110,6 +114,13 @@ def check_multiple_genotypes(contigs_csv, read_count):
                      genotypes)
         return DD.FAIL if genotype_count > 2 else DD.PASS
 
+    @staticmethod
+    def check_type_error(_contigs_csv, read_count, exception):
+        logger.debug('Result: %s exception from %d reads.',
+                     exception,
+                     read_count)
+        return DD.FAIL if isinstance(exception, TypeError) else DD.PASS
+
     def write_simple_fastq(self, filename1, filename2, read_indexes):
         selected_reads = (self.reads[i] for i in read_indexes)
         with open(filename1, 'w') as f1, open(filename2, 'w') as f2: