Fix variable shadowing and update instance method calls across codebase

- Fix samtools variable shadowing in bwa.py, novoalign.py, picard.py, reports.py
- Fix bwa variable shadowing in reports.py
- Fix picard variable shadowing in kb.py, kma.py, kraken2.py
- Update minimap2.py to use samtools_tool instance for all method calls
- Remove duplicate imports in minimap2.py and bwa.py
- Fix kraken2 fixture shadowing in test_integration_kraken2.py
- Pin muscle=3.8.1551 in phylo.txt to avoid MUSCLE v5 CLI incompatibility

The pattern `var = module.Class()` shadows the module import, causing
UnboundLocalError when the module is referenced later in the function.
Fixed by renaming to `var_tool = module.Class()`.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

Author: dpark01

docker/requirements/phylo.txt

@@ -3,6 +3,6 @@ bamtools>=2.5.2
 lofreq>=2.1.5
 mafft>=7.508
 mummer4>=4.0.0rc1
-muscle>=3.8
+muscle=3.8.1551
 snpeff>=4.3.1t
 vphaser2>=2.0

src/viral_ngs/classify/kb.py

@@ -173,12 +173,12 @@ def classify(self, in_bam, index_file, out_dir, t2g_file, k=31, parity='single',
 
                 # Do not convert this to samtools bam2fq unless we can figure out how to replicate
                 # the clipping functionality of Picard SamToFastq
-                picard = picard.SamToFastqTool()
+                picard_tool = picard.SamToFastqTool()
                 picard_opts = {
                     'CLIPPING_ATTRIBUTE': picard.SamToFastqTool.illumina_clipping_attribute,
                     'CLIPPING_ACTION': 'X'
                 }
-                picard.execute(in_bam, tmp_fastq1, tmp_fastq2, outFastq0=tmp_fastq3,
+                picard_tool.execute(in_bam, tmp_fastq1, tmp_fastq2, outFastq0=tmp_fastq3,
                             picardOptions=picard.PicardTools.dict_to_picard_opts(picard_opts),
                             JVMmemory=picard.jvmMemDefault)
 
@@ -264,12 +264,12 @@ def extract(self, in_bam, index_file, target_ids, out_dir, t2g_file, protein=Fal
 
                 # Do not convert this to samtools bam2fq unless we can figure out how to replicate
                 # the clipping functionality of Picard SamToFastq
-                picard = picard.SamToFastqTool()
+                picard_tool = picard.SamToFastqTool()
                 picard_opts = {
                     'CLIPPING_ATTRIBUTE': picard.SamToFastqTool.illumina_clipping_attribute,
                     'CLIPPING_ACTION': 'X'
                 }
-                picard.execute(in_bam, tmp_fastq1, tmp_fastq2, outFastq0=tmp_fastq3,
+                picard_tool.execute(in_bam, tmp_fastq1, tmp_fastq2, outFastq0=tmp_fastq3,
                             picardOptions=picard.PicardTools.dict_to_picard_opts(picard_opts),
                             JVMmemory=picard.jvmMemDefault)
 

src/viral_ngs/classify/kma.py

@@ -104,12 +104,12 @@ def classify(self, in_bam, db, out_prefix, num_threads=None):
         tmp_fastq3 = file.mkstempfname('.s.fastq')
 
         try:
-            picard = picard.SamToFastqTool()
+            picard_tool = picard.SamToFastqTool()
             picard_opts = {
                 'CLIPPING_ATTRIBUTE': picard.SamToFastqTool.illumina_clipping_attribute,
                 'CLIPPING_ACTION': 'X'
             }
-            picard.execute(in_bam, tmp_fastq1, tmp_fastq2, outFastq0=tmp_fastq3,
+            picard_tool.execute(in_bam, tmp_fastq1, tmp_fastq2, outFastq0=tmp_fastq3,
                            picardOptions=picard.PicardTools.dict_to_picard_opts(picard_opts),
                            JVMmemory=picard.jvmMemDefault)
 

src/viral_ngs/classify/kraken2.py

@@ -201,12 +201,12 @@ def classify(self, in_bam, db, out_reads=None, out_report=None,
         tmp_fastq3 = file.mkstempfname('.s.fastq')
         # Do not convert this to samtools bam2fq unless we can figure out how to replicate
         # the clipping functionality of Picard SamToFastq
-        picard = picard.SamToFastqTool()
+        picard_tool = picard.SamToFastqTool()
         picard_opts = {
             'CLIPPING_ATTRIBUTE': picard.SamToFastqTool.illumina_clipping_attribute,
             'CLIPPING_ACTION': 'X'
         }
-        picard.execute(in_bam, tmp_fastq1, tmp_fastq2, outFastq0=tmp_fastq3,
+        picard_tool.execute(in_bam, tmp_fastq1, tmp_fastq2, outFastq0=tmp_fastq3,
                        picardOptions=picard.PicardTools.dict_to_picard_opts(picard_opts),
                        JVMmemory=picard.jvmMemDefault)
 

src/viral_ngs/core/bwa.py

@@ -11,7 +11,6 @@
 import shutil
 import concurrent.futures
 
-from . import samtools, picard  # was: from viral_ngs import tools
 from . import samtools
 from . import picard
 from . import file as util_file, misc as util_misc  # was: from viral_ngs import util
@@ -58,11 +57,11 @@ def align_mem_bam(self, inBam, refDb, outBam, options=None,
                       min_score_to_filter=None, threads=None, JVMmemory=None, invert_filter=False, should_index=True):
         options = options or []
 
-        samtools = samtools.SamtoolsTool()
+        samtools_tool = samtools.SamtoolsTool()
         threads = util_misc.sanitize_thread_count(threads)
 
         # fetch list of RGs
-        rgs = list(samtools.getReadGroups(inBam).keys())
+        rgs = list(samtools_tool.getReadGroups(inBam).keys())
 
         if len(rgs) == 0:
             # Can't do this
@@ -138,10 +137,10 @@ def align_mem_one_rg(self, inBam, refDb, outBam, rgid=None, options=None,
         """
         options = options or []
 
-        samtools = samtools.SamtoolsTool()
+        samtools_tool = samtools.SamtoolsTool()
 
         # Require exactly one RG
-        rgs = samtools.getReadGroups(inBam)
+        rgs = samtools_tool.getReadGroups(inBam)
         if len(rgs) == 0:
             raise InvalidBamHeaderError("{} lacks read groups".format(inBam))
         elif len(rgs) == 1:
@@ -157,27 +156,27 @@ def align_mem_one_rg(self, inBam, refDb, outBam, rgid=None, options=None,
         removeInput = False
         if len(rgs) == 1:
             one_rg_inBam = inBam
-            samtools.SamtoolsTool().dumpHeader(one_rg_inBam, headerFile)
+            samtools_tool.dumpHeader(one_rg_inBam, headerFile)
         else:
             # strip inBam to one read group
             with util_file.tempfname('.onebam.bam') as tmp_bam:
-                samtools.view(['-b', '-r', rgid], inBam, tmp_bam)
+                samtools_tool.view(['-b', '-r', rgid], inBam, tmp_bam)
                 # special exit if this file is empty
-                if samtools.count(tmp_bam) == 0:
+                if samtools_tool.count(tmp_bam) == 0:
                     log.warning("No reads present for RG %s in file: %s", rgid, inBam)
                     return
                 # simplify BAM header otherwise Novoalign gets confused
                 one_rg_inBam = util_file.mkstempfname('.{}.in.bam'.format(rgid))
                 removeInput = True
-                
+
                 with open(headerFile, 'wt') as outf:
-                    for row in samtools.getHeader(inBam):
+                    for row in samtools_tool.getHeader(inBam):
                         if len(row) > 0 and row[0] == '@RG':
                             if rgid != list(x[3:] for x in row if x.startswith('ID:'))[0]:
                                 # skip all read groups that are not rgid
                                 continue
                         outf.write('\t'.join(row) + '\n')
-                samtools.reheader(tmp_bam, headerFile, one_rg_inBam)
+                samtools_tool.reheader(tmp_bam, headerFile, one_rg_inBam)
 
         # perform actual alignment
 
@@ -212,10 +211,10 @@ def mem(self, inReads, refDb, outAlign, options=None, min_score_to_filter=None,
         if '-t' not in options:
             options.extend(('-t', str(threads)))
 
-        samtools = samtools.SamtoolsTool()
+        samtools_tool = samtools.SamtoolsTool()
 
         aln_sam = util_file.mkstempfname('.aligned.sam')
-        fastq_pipe = samtools.bam2fq_pipe(inReads)
+        fastq_pipe = samtools_tool.bam2fq_pipe(inReads)
         self.execute('mem', options + ['-p', refDb, '-'], stdout=aln_sam, stdin=fastq_pipe.stdout)
 
         if fastq_pipe.poll():
@@ -231,12 +230,12 @@ def mem(self, inReads, refDb, outAlign, options=None, min_score_to_filter=None,
             aln_sam_filtered = aln_sam
 
 
-        samtools.sort(aln_sam_filtered, outAlign, threads=threads)
+        samtools_tool.sort(aln_sam_filtered, outAlign, threads=threads)
         os.unlink(aln_sam_filtered)
 
         # cannot index sam files; only do so if a bam/cram is desired
         if should_index and (outAlign.endswith(".bam") or outAlign.endswith(".cram")):
-            samtools.index(outAlign)
+            samtools_tool.index(outAlign)
 
     def filter_sam_on_alignment_score(self, in_sam, out_sam, min_score_to_filter,
                                       bwa_options, invert_filter=False):

src/viral_ngs/core/minimap2.py

@@ -12,7 +12,6 @@
 
 from Bio import SeqIO
 
-from . import samtools, picard  # was: from viral_ngs import tools
 from . import samtools
 from . import picard
 from . import file as util_file, misc as util_misc  # was: from viral_ngs import util
@@ -54,7 +53,7 @@ def align_bam(self, inBam, refDb, outBam, options=None,
         threads = util_misc.sanitize_thread_count(threads)
 
         # fetch list of RGs
-        rgs = list(samtools.getReadGroups(inBam).keys())
+        rgs = list(samtools_tool.getReadGroups(inBam).keys())
 
         if len(rgs) == 0:
             # Can't do this
@@ -79,16 +78,16 @@ def align_bam(self, inBam, refDb, outBam, options=None,
                     threads=threads,
                     should_index=False  # Don't index intermediate BAMs that will be merged
                 )
-                if not samtools.isEmpty(tmp_bam):
+                if not samtools_tool.isEmpty(tmp_bam):
                     align_bams.append(tmp_bam)
                 else:
                     log.warning("No alignment output for RG %s in file %s against %s", rg, inBam, refDb)
 
             if len(align_bams)==0:
                 log.warning("All read groups in file %s appear to be empty.", inBam)
                 with util_file.tempfname('.empty.sam') as empty_sam:
-                    samtools.dumpHeader(inBam, empty_sam)
-                    samtools.sort(empty_sam, outBam)
+                    samtools_tool.dumpHeader(inBam, empty_sam)
+                    samtools_tool.sort(empty_sam, outBam)
             else:
                 # Merge BAMs, sort, and index
                 picardOptions = ['SORT_ORDER=coordinate', 'USE_THREADING=true', 'CREATE_INDEX=true']
@@ -117,7 +116,7 @@ def align_one_rg(self, inBam, refDb, outBam, rgid=None, preset=None, options=Non
         samtools_tool = samtools.SamtoolsTool()
 
         # Require exactly one RG
-        rgs = samtools.getReadGroups(inBam)
+        rgs = samtools_tool.getReadGroups(inBam)
         if len(rgs) == 0:
             raise InvalidBamHeaderError("{} lacks read groups".format(inBam))
         elif len(rgs) == 1:
@@ -133,13 +132,13 @@ def align_one_rg(self, inBam, refDb, outBam, rgid=None, preset=None, options=Non
         removeInput = False
         if len(rgs) == 1:
             one_rg_inBam = inBam
-            samtools.SamtoolsTool().dumpHeader(one_rg_inBam, headerFile)
+            samtools_tool.dumpHeader(one_rg_inBam, headerFile)
         else:
             # strip inBam to one read group
             with util_file.tempfname('.onebam.bam') as tmp_bam:
-                samtools.view(['-1', '-r', rgid], inBam, tmp_bam)
+                samtools_tool.view(['-1', '-r', rgid], inBam, tmp_bam)
                 # special exit if this file is empty
-                if samtools.isEmpty(tmp_bam):
+                if samtools_tool.isEmpty(tmp_bam):
                     log.warning("No reads present for RG %s in file: %s", rgid, inBam)
                     shutil.copyfile(tmp_bam, outBam)
                     return
@@ -148,13 +147,13 @@ def align_one_rg(self, inBam, refDb, outBam, rgid=None, preset=None, options=Non
                 removeInput = True
 
                 with open(headerFile, 'wt') as outf:
-                    for row in samtools.getHeader(inBam):
+                    for row in samtools_tool.getHeader(inBam):
                         if len(row) > 0 and row[0] == '@RG':
                             if rgid != list(x[3:] for x in row if x.startswith('ID:'))[0]:
                                 # skip all read groups that are not rgid
                                 continue
                         outf.write('\t'.join(row) + '\n')
-                samtools.reheader(tmp_bam, headerFile, one_rg_inBam)
+                samtools_tool.reheader(tmp_bam, headerFile, one_rg_inBam)
 
         # get the read group line to give to mm2
         readgroup_line = ""
@@ -187,7 +186,7 @@ def align_one_rg(self, inBam, refDb, outBam, rgid=None, preset=None, options=Non
             options.extend(('-x', preset))
 
         # perform actual alignment
-        if samtools.isEmpty(one_rg_inBam):
+        if samtools_tool.isEmpty(one_rg_inBam):
             log.warning("Input file %s appears to lack reads for RG '%s'", inBam, rgid)
             # minimap doesn't like empty inputs, so copy empty bam through
             # samtools.sort(one_rg_inBam, outBam)
@@ -213,20 +212,21 @@ def align_cmd(self, inReads, refDb, outAlign, options=None, threads=None, should
         samtools_tool = samtools.SamtoolsTool()
 
         with util_file.tempfname('.aligned.sam') as aln_sam:
-            fastq_pipe = samtools.bam2fq_pipe(inReads)
+            fastq_pipe = samtools_tool.bam2fq_pipe(inReads)
             options.extend(('-a', refDb, '-', '-o', aln_sam))
             self.execute(options, stdin=fastq_pipe.stdout)
             if fastq_pipe.wait():
                 raise subprocess.CalledProcessError(fastq_pipe.returncode, "samtools.bam2fq_pipe() for {}".format(inReads))
-            samtools.sort(aln_sam, outAlign, threads=threads)
+            samtools_tool.sort(aln_sam, outAlign, threads=threads)
 
         # cannot index sam files; only do so if a bam/cram is desired
         if should_index and (outAlign.endswith(".bam") or outAlign.endswith(".cram")):
-            samtools.index(outAlign, threads=threads)
+            samtools_tool.index(outAlign, threads=threads)
 
     def scaffold(self, contigs_fasta, ref_fasta, outAlign, divergence=20, options=None, threads=None):
         options = [] if not options else options
 
+        samtools_tool = samtools.SamtoolsTool()
         threads = util_misc.sanitize_thread_count(threads)
         if '-t' not in options:
             options.extend(('-t', str(threads)))
@@ -243,11 +243,11 @@ def scaffold(self, contigs_fasta, ref_fasta, outAlign, divergence=20, options=No
         with util_file.tempfname('.aligned.sam') as aln_sam:
             options.extend(('-a', ref_fasta, contigs_fasta, '-o', aln_sam))
             self.execute(options)
-            samtools.sort(aln_sam, outAlign, threads=threads)
+            samtools_tool.sort(aln_sam, outAlign, threads=threads)
 
         # cannot index sam files; only do so if a bam/cram is desired
         if (outAlign.endswith(".bam") or outAlign.endswith(".cram")):
-            samtools.index(outAlign)
+            samtools_tool.index(outAlign)
 
     def idxstats(self, inReads, refDb, outIdxstats, outReadlist=None, threads=None):
         """
@@ -302,7 +302,7 @@ def idxstats(self, inReads, refDb, outIdxstats, outReadlist=None, threads=None):
 
         try:
             # Start samtools bam2fq pipe for input (use 4 threads for BAM decompression)
-            fastq_pipe = samtools.bam2fq_pipe(inReads, threads=4)
+            fastq_pipe = samtools_tool.bam2fq_pipe(inReads, threads=4)
 
             # Start minimap2 process with PAF output to stdout
             tool_cmd = [self.install_and_get_path()] + options

src/viral_ngs/core/novoalign.py

@@ -77,12 +77,12 @@ def execute(self, inBam, refFasta, outBam, options=None, min_qual=0, JVMmemory=N
         '''
         options = options or ["-r", "Random"]
 
-        samtools = samtools.SamtoolsTool()
+        samtools_tool = samtools.SamtoolsTool()
 
         # fetch list of RGs
-        rgs = samtools.getReadGroups(inBam)
+        rgs = samtools_tool.getReadGroups(inBam)
 
-        rgs_list = list(samtools.getReadGroups(inBam).keys())
+        rgs_list = list(samtools_tool.getReadGroups(inBam).keys())
         if len(rgs) == 0:
             # Can't do this
             raise InvalidBamHeaderError("{} lacks read groups".format(inBam))
@@ -127,10 +127,10 @@ def align_one_rg_bam(self, inBam, refFasta, outBam, rgid=None, rgs=None, options
         '''
         options = options or ["-r", "Random"]
 
-        samtools = samtools.SamtoolsTool()
+        samtools_tool = samtools.SamtoolsTool()
 
         # Require exactly one RG
-        rgs = rgs if rgs is not None else samtools.getReadGroups(inBam)
+        rgs = rgs if rgs is not None else samtools_tool.getReadGroups(inBam)
         if len(rgs) == 0:
             raise InvalidBamHeaderError("{} lacks read groups".format(inBam))
         elif len(rgs) == 1:
@@ -148,22 +148,22 @@ def align_one_rg_bam(self, inBam, refFasta, outBam, rgid=None, rgs=None, options
         else:
             # strip inBam to one read group
             tmp_bam = util_file.mkstempfname('.onebam.bam')
-            samtools.view(['-b', '-r', rgid], inBam, tmp_bam)
+            samtools_tool.view(['-b', '-r', rgid], inBam, tmp_bam)
             # special exit if this file is empty
-            if samtools.count(tmp_bam) == 0:
+            if samtools_tool.count(tmp_bam) == 0:
                 return
 
             # simplify BAM header otherwise Novoalign gets confused
             one_rg_inBam = util_file.mkstempfname('.{}.in.bam'.format(rgid))
             headerFile = util_file.mkstempfname('.{}.header.txt'.format(rgid))
             with open(headerFile, 'wt') as outf:
-                for row in samtools.getHeader(inBam):
+                for row in samtools_tool.getHeader(inBam):
                     if len(row) > 0 and row[0] == '@RG':
                         if rgid != list(x[3:] for x in row if x.startswith('ID:'))[0]:
                             # skip all read groups that are not rgid
                             continue
                     outf.write('\t'.join(row) + '\n')
-            samtools.reheader(tmp_bam, headerFile, one_rg_inBam)
+            samtools_tool.reheader(tmp_bam, headerFile, one_rg_inBam)
             os.unlink(tmp_bam)
             os.unlink(headerFile)
 
@@ -179,7 +179,7 @@ def align_one_rg_bam(self, inBam, refFasta, outBam, rgid=None, rgs=None, options
         # Samtools filter (optional)
         if min_qual:
             tmp_bam2 = util_file.mkstempfname('.filtered.bam')
-            samtools.view(['-b', '-S', '-1', '-q', str(min_qual)], tmp_sam, tmp_bam2)
+            samtools_tool.view(['-b', '-S', '-1', '-q', str(min_qual)], tmp_sam, tmp_bam2)
             os.unlink(tmp_sam)
             tmp_sam = tmp_bam2
 

src/viral_ngs/core/picard.py

@@ -389,8 +389,8 @@ def downsample_to_approx_count(
         JVMmemory=None
     ):    # pylint: disable=W0221):
 
-        samtools = samtools.SamtoolsTool()
-        total_read_count = samtools.count(inBam)
+        samtools_tool = samtools.SamtoolsTool()
+        total_read_count = samtools_tool.count(inBam)
 
         if total_read_count == 0:
             _log.info("Input BAM has no reads. Copying to output.")