Fix bugs related to lowercase bases

test/test_misc.py

@@ -199,11 +199,20 @@ def test_load_fasta_4():
     assert seqs[1][1].startswith('ATTCTCAGAATGGCGTAG')
 
 
+def test_load_fasta_5():
+    seqs = trycycler.misc.load_fasta('test/test_misc/lowercase.fasta')
+    assert len(seqs) == 2
+    assert seqs[0][0] == 'A'
+    assert seqs[0][1].startswith('TTGCCTGTAGTCGGGACC')
+    assert seqs[1][0] == 'B'
+    assert seqs[1][1].startswith('ATTCTCAGAATGGCGTAG')
+
+
 def test_get_default_thread_count():
     assert 1 <= trycycler.misc.get_default_thread_count() <= 16
 
 
-def test_write_seq_to_fasta():
+def test_write_seq_to_fasta_1():
     with tempfile.TemporaryDirectory() as temp_dir:
         filename = pathlib.Path(temp_dir) / 'temp.fasta'
         trycycler.misc.write_seq_to_fasta('CAGAATGGCGT', 'name', filename)
@@ -213,6 +222,17 @@ def test_write_seq_to_fasta():
         assert seqs[0][1] == 'CAGAATGGCGT'
 
 
+def test_write_seq_to_fasta_2():
+    # Same test, but with lowercase bases in input (should be made uppercase in saved file).
+    with tempfile.TemporaryDirectory() as temp_dir:
+        filename = pathlib.Path(temp_dir) / 'temp.fasta'
+        trycycler.misc.write_seq_to_fasta('CAgaaTgGcgt', 'name', filename)
+        seqs = trycycler.misc.load_fasta(filename)
+        assert len(seqs) == 1
+        assert seqs[0][0] == 'name'
+        assert seqs[0][1] == 'CAGAATGGCGT'
+
+
 def test_reverse_complement_1():
     assert trycycler.misc.reverse_complement('GGGGaaaaaaaatttatatat') == 'atatataaattttttttCCCC'
 

test/test_misc/lowercase.fasta

@@ -0,0 +1,4 @@
+>A info
+ttgcctgtagtcgggaccccgtgactaggaaagcaatcagcgactaacaggcggagaccgtctatagcgcacggggtgtagttggctattactgatctct
+>B stuff
+attctcagaatggcgtagtattcatatttgttcgtagcccgcctccgtacatgttattgtgctcatcggtggcctgcgccgtggggagtgcaaaacgtgg

test/test_msa.py

@@ -126,3 +126,17 @@ def test_msa_3():
 
     assert aligned_seqs == [('A', 'ATGTAAAGGTTCCGGGGCACTTAGCAGCTCCACAAATCCATT-CCAACCTATA'),
                             ('B', 'ATGTAAAGGTT--GGGGCACTTAGCA-CTCCACACATCCATTGCCAACCTATA')]
+
+
+def test_msa_4():
+    # Same sequences, but with lower case bases (should be made uppercase in alignment).
+    seqs = [('A', 'ATGtAAAGGTTcCGGGGCACttAGCaGCTCCACAaAtCcATTCCAACcTaTA'),
+            ('B', 'ATGTaAaGGtTgGgGCAcTTAGCACTCCaCACATcCAttGCCaACCTATA')]
+    with tempfile.TemporaryDirectory() as temp_dir:
+        temp_dir = pathlib.Path(temp_dir)
+        save_input_sequences(temp_dir, seqs)
+        trycycler.msa.msa(create_args(temp_dir, kmer=32, step=1000, lookahead=10000, threads=8))
+        aligned_seqs = load_fasta(temp_dir / '3_msa.fasta')
+
+    assert aligned_seqs == [('A', 'ATGTAAAGGTTCCGGGGCACTTAGCAGCTCCACAAATCCATT-CCAACCTATA'),
+                            ('B', 'ATGTAAAGGTT--GGGGCACTTAGCA-CTCCACACATCCATTGCCAACCTATA')]

trycycler/cluster.py

@@ -324,7 +324,7 @@ def complete_linkage(seqs, seq_names, depths, distances, threshold, out_dir):
             seq_depth = depths[name]
             with open(seq_fasta, 'wt') as f:
                 f.write(f'>{name}\n')
-                f.write(f'{seq}\n')
+                f.write(f'{seq.upper()}\n')
             log_lines.append((seq_fasta, seq_length, seq_depth))
 
         for seq_fasta, seq_length, seq_depth in log_lines:

trycycler/consensus.py

@@ -681,7 +681,7 @@ def save_seqs_to_fasta(seqs, filename, extra_newline=True):
     with open(filename, 'wt') as fasta:
         for name, seq in seqs.items():
             fasta.write(f'>{name}\n')
-            fasta.write(f'{seq}\n')
+            fasta.write(f'{seq.upper()}\n')
     if extra_newline:
         log()
 

trycycler/misc.py

@@ -138,7 +138,7 @@ def load_fasta(fasta_filename, include_full_header=False):
                     sequence = []
                 name = line[1:]
             else:
-                sequence.append(line)
+                sequence.append(line.upper())
         if name:
             if include_full_header:
                 fasta_seqs.append((name.split()[0], name, ''.join(sequence)))
@@ -154,7 +154,7 @@ def get_default_thread_count():
 def write_seq_to_fasta(seq, name, filename):
     with open(filename, 'wt') as f:
         f.write(f'>{name}\n')
-        f.write(f'{seq}\n')
+        f.write(f'{seq.upper()}\n')
 
 
 REV_COMP_DICT = {'A': 'T', 'T': 'A', 'G': 'C', 'C': 'G', 'a': 't', 't': 'a', 'g': 'c', 'c': 'g',

trycycler/msa.py

@@ -166,7 +166,7 @@ def merge_pieces(temp_dir: pathlib.Path, cluster_dir, seqs):
     for f in msa_fasta_files:
         parts = dict(load_fasta(f))
         for n in seq_names:
-            aligned_seq_parts[n].append(parts[n])
+            aligned_seq_parts[n].append(parts[n].upper())
     aligned_seqs = {}
     for n in seq_names:
         aligned_seqs[n] = ''.join(aligned_seq_parts[n])
@@ -184,7 +184,7 @@ def merge_pieces(temp_dir: pathlib.Path, cluster_dir, seqs):
     # Sanity check: the MSA sequences should match the original sequences.
     for n in seq_names:
         msa_minus_dashes = final_seqs[n].replace('-', '')
-        assert seqs[n] == msa_minus_dashes
+        assert seqs[n].upper() == msa_minus_dashes
 
     # Save the full MSA to file.
     final_msa_fasta_filename = cluster_dir / '3_msa.fasta'

trycycler/reconcile.py

@@ -126,7 +126,7 @@ def save_seqs_to_fasta(seqs, filename):
     with open(filename, 'wt') as fasta:
         for name, seq in seqs.items():
             fasta.write(f'>{name}\n')
-            fasta.write(f'{seq}\n')
+            fasta.write(f'{seq.upper()}\n')
     log()