Merge pull request #64 from ShawHahnLab/release-0.5.1

Version 0.5.1

CHANGELOG.md

@@ -1,5 +1,29 @@
 # Changelog
 
+## 0.5.1 - 2023-03-31
+
+### Changed
+
+ * `convert` now handles edge cases for sequence input and tabular output by
+   always including a sequence description column in the output ([#63])
+
+### Fixed
+
+ * `msa` will now bypass calling MUSCLE when called with just a single input
+   sequence, avoiding a crash ([#62])
+ * `convert` will now obey a custom sequence description column name if one is
+   given with `--col-seq-desc` ([#60])
+ * `getreads` command is now compatible with the latest available version of
+   bcl2fastq, v2.20.0.422 ([#58])
+ * `tree` command can now handle assigning a color code when exactly one
+   sequence set is defined ([#57])
+
+[#63]: https://github.com/ShawHahnLab/igseq/pull/63
+[#62]: https://github.com/ShawHahnLab/igseq/pull/62
+[#60]: https://github.com/ShawHahnLab/igseq/pull/60
+[#58]: https://github.com/ShawHahnLab/igseq/pull/58
+[#57]: https://github.com/ShawHahnLab/igseq/pull/57
+
 ## 0.5.0 - 2023-01-04
 
 ### Added

conda/meta.yaml

@@ -1,5 +1,5 @@
 # https://docs.conda.io/projects/conda-build/en/latest/resources/define-metadata.html
-{% set version = "0.5.0" %}
+{% set version = "0.5.1" %}
 {% set build = "0" %}
 
 package:

igseq/convert.py

@@ -20,4 +20,12 @@ def convert(path_in, path_out, fmt_in=None, fmt_out=None, colmap=None, dummyqual
     with RecordReader(path_in, fmt_in, colmap, dry_run=dry_run) as reader, \
         RecordWriter(path_out, fmt_out, colmap, dummyqual=dummyqual, dry_run=dry_run) as writer:
         for record in reader:
+            # special case for descriptions: they may or may not exist on any
+            # particular record for seq input, but for tabular output, we have
+            # to have consistent columns.  So in that case make sure to include
+            # a description column by forcing one for the first record to be
+            # written.
+            if not writer.writer and writer.tabular and not reader.tabular:
+                key = reader.colmap["sequence_description"]
+                record[key] = record.get(key, "")
             writer.write(record)

igseq/data/examples/outputs/convert/unwrapped.csv

@@ -1,2 +1,2 @@
-sequence_id,sequence
-seqid,ACTGACTGACTGACTG
+sequence_id,sequence,sequence_description
+seqid,ACTGACTGACTGACTG,

igseq/getreads.py

@@ -15,6 +15,7 @@
 those options.
 """
 
+import sys
 import logging
 import subprocess
 from tempfile import NamedTemporaryFile
@@ -98,10 +99,6 @@ def getreads(path_input, dir_out, path_counts="", extra_args=None, threads_load=
                 "--sample-sheet", sample_sheet.name,
                 # parallel processing during loading can help a bit in my tests
                 "--loading-threads", threads_load,
-                # parallel processing does *not* help during the bcl2fastq
-                # demultiplexing step, go figure, when we don't have any
-                # demultiplexing to perform here
-                "--demultiplexing-threads", 1,
                 # parallel processing in the processing step helps quite a bit
                 "--processing-threads", threads_proc,
                 # help text says "this must not be higher than number of
@@ -165,4 +162,6 @@ def _run_bcl2fastq(args, extra_args=None):
             raise util.IgSeqError(f"bcl2fastq arg collision from extra arguments: {shared}")
         args += extra_args
     LOGGER.info("bcl2fastq command: %s", args)
-    subprocess.run(args, check=True)
+    proc = subprocess.run(args, check=True, capture_output=True, text=True)
+    sys.stdout.write(proc.stdout)
+    sys.stderr.write(proc.stderr)

igseq/identity.py

@@ -11,7 +11,7 @@
 
 The scoring is based on a simple global pairwise alignment, with matches scored
 as 1, mismatches and gaps 0.  Any existing gaps are removed before comparing
-sequences, and differeces in case (lower/upper) are disregarded.
+sequences, and differences in case (lower/upper) are disregarded.
 """
 
 import logging

igseq/igblast.py

@@ -10,7 +10,7 @@
 command, so you can configure things like the output format and file path.  See
 igblastn -help for those options.  Any igblastn argument can be given with two
 dashes if needed to force igseq to handle it correctly (for example,
--num_alignments_V will be interprted as -n um_alignments_V, but
+-num_alignments_V will be interpreted as -n um_alignments_V, but
 --num_alignments_V will work).
 """
 

igseq/msa.py

@@ -37,8 +37,9 @@ def msa(path_in, path_out, fmt_in=None, fmt_out=None, colmap=None, dry_run=False
 def run_muscle(records):
     """Align a set of records with MUSCLE."""
     # muscle crashes with empty input, so we'll just do a noop for that case
-    if not records:
-        LOGGER.warning("no records provided to align; skipping MUSCLE")
+    if len(records) < 2:
+        detail = "only one record" if len(records) else "no records"
+        LOGGER.warning("%s provided to align; skipping MUSCLE", detail)
         return records
     args = ["muscle", "-align", "/dev/stdin", "-output", "/dev/stdout"]
     with Popen(args, stdin=PIPE, stdout=PIPE, stderr=PIPE, text=True) as proc:

igseq/record.py

@@ -94,6 +94,10 @@ def infer_fmt(self, fmt=None):
             fmt = fmt_inferred
         return fmt
 
+    @property
+    def tabular(self):
+        return self.fmt in ["csv", "tsv", "csvgz", "tsvgz"]
+
     @staticmethod
     def _infer_fmt(path):
         try:
@@ -141,7 +145,7 @@ def decode_record(self, obj):
             if quals:
                 record[self.colmap["sequence_quality"]] = self.encode_phred(quals)
             if seq_desc is not None:
-                record["sequence_description"] = seq_desc
+                record[self.colmap["sequence_description"]] = seq_desc
         else:
             record = obj
         return record
@@ -267,7 +271,7 @@ def _write_fa(self, record):
         seq = record[self.colmap["sequence"]]
         defline = record[self.colmap["sequence_id"]]
         desc = record.get(self.colmap["sequence_description"])
-        if desc is not None:
+        if desc:
             defline += f" {desc}"
         if not self.dry_run:
             self.handle.write(f">{defline}\n{seq}\n")
@@ -283,7 +287,7 @@ def _write_fq(self, record):
                 "No quality scores available, using default dummy value: %s",
                 DEFAULT_DUMMY_QUAL)
             quals = "".join(DEFAULT_DUMMY_QUAL * len(seq))
-        if desc is not None:
+        if desc:
             defline += f" {desc}"
         if not self.dry_run:
             self.handle.write(f"@{defline}\n{seq}\n+\n{quals}\n")

igseq/tree.py

@@ -314,7 +314,7 @@ def make_seq_set_colors(seq_sets):
         # adapted from SONAR
         # this stretches across COLORS in even increments for as many as we need here
         num = len(colors.COLORS)
-        subset = [int( a * (num-1) / (len(seq_sets)-1) ) for a in range(num)]
+        subset = [int( a * (num-1) / max(1, (len(seq_sets)-1)) ) for a in range(num)]
         try:
             seq_set_colors[set_name] = colors.color_str_to_trio(colors.COLORS[subset[idx]])
         except IndexError:

igseq/trim.py

@@ -4,7 +4,7 @@
 By default this will remove any instances of the R2 adapter found toward the
 end of R1 and the R1 adapter found toward the end of R2.  It will also insist
 that the 5' RACE Anchor be found at the start of R1, discarding read pairs that
-are mising the anchor.  The adapter sequences will be determined from the
+are missing the anchor.  The adapter sequences will be determined from the
 barcodes used for each sample and the selected species.
 """
 

igseq/version.py

@@ -4,4 +4,4 @@
 See https://www.python.org/dev/peps/pep-0396/
 """
 
-__version__ = "0.5.0"
+__version__ = "0.5.1"

test_igseq/data/test_convert/TestConvert/unwrapped.csv

@@ -1,2 +1,2 @@
-sequence_id,sequence
-seqid,ACTGACTGACTGACTG
+sequence_id,sequence,sequence_description
+seqid,ACTGACTGACTGACTG,

test_igseq/data/test_convert/TestConvert/unwrapped.tsv

@@ -1,2 +1,2 @@
-sequence_id	sequence
-seqid	ACTGACTGACTGACTG
+sequence_id	sequence	sequence_description
+seqid	ACTGACTGACTGACTG	

test_igseq/data/test_convert/TestConvert/unwrapped_quals.csv

@@ -1,2 +1,2 @@
-sequence_id,sequence,sequence_quality
-seqid,ACTGACTGACTGACTG,IIIIIIIIIIIIIIII
+sequence_id,sequence,sequence_quality,sequence_description
+seqid,ACTGACTGACTGACTG,IIIIIIIIIIIIIIII,

test_igseq/data/test_convert/TestConvert/unwrapped_quals.tsv

@@ -1,2 +1,2 @@
-sequence_id	sequence	sequence_quality
-seqid	ACTGACTGACTGACTG	IIIIIIIIIIIIIIII
+sequence_id	sequence	sequence_quality	sequence_description
+seqid	ACTGACTGACTGACTG	IIIIIIIIIIIIIIII	

test_igseq/data/test_convert/TestConvertCustomCols/unwrapped.alt1.csv

@@ -1,2 +1,2 @@
-SeqID,sequence
-seqid,ACTGACTGACTGACTG
+SeqID,sequence,sequence_description
+seqid,ACTGACTGACTGACTG,

test_igseq/data/test_convert/TestConvertCustomCols/unwrapped.alt2.csv

@@ -1,2 +1,2 @@
-sequence_id,sequence
-seqid,ACTGACTGACTGACTG
+sequence_id,sequence,sequence_description
+seqid,ACTGACTGACTGACTG,

test_igseq/data/test_convert/TestConvertCustomCols/unwrapped.csv

@@ -1,2 +1,2 @@
-SeqID,Seq
-seqid,ACTGACTGACTGACTG
+SeqID,Seq,sequence_description
+seqid,ACTGACTGACTGACTG,

test_igseq/data/test_convert/TestConvertCustomCols/unwrapped_desc.csv

@@ -0,0 +1,2 @@
+SeqID,Seq,SeqDesc
+seqid,ACTGACTGACTGACTG,desc

test_igseq/data/test_convert/TestConvertCustomCols/unwrapped_desc_alt.csv

@@ -0,0 +1,3 @@
+SeqID,Seq,SeqDesc
+seqid,ACTGACTGACTGACTG,
+seqid2,TCTGACTCACTGAGTA,desc

test_igseq/data/test_convert/TestConvertCustomCols/unwrapped_quals.csv

@@ -1,2 +1,2 @@
-SeqID,Seq,SeqQual
-seqid,ACTGACTGACTGACTG,IIIIIIIIIIIIIIII
+SeqID,Seq,SeqQual,sequence_description
+seqid,ACTGACTGACTGACTG,IIIIIIIIIIIIIIII,

test_igseq/data/test_convert/TestConvertCustomCols/wrapped_desc.fasta

@@ -0,0 +1,3 @@
+>seqid desc
+ACTGACTG
+ACTGACTG

test_igseq/data/test_convert/TestConvertCustomCols/wrapped_desc_alt.fasta

@@ -0,0 +1,6 @@
+>seqid
+ACTGACTG
+ACTGACTG
+>seqid2 desc
+TCTGACTC
+ACTGAGTA