A couple of changes related to filetypes

Author: joshfactorial

neat/bacterial_wrapper/runner.py

@@ -125,14 +125,16 @@ def bacterial_wrapper(reference_file, bacteria_name, ref_config_file, output_dir
 
 # Stitching all outputs together - Keshav's script
 
-def concat_fq(input_files: List[Path], dest: BgzfWriter) -> None:
+def concat_fq(input_files: List[Path], dest: Path) -> None:
 
     if not input_files:
+        # Nothing to do, and no error to throw
         return
-    
-    for input_file in input_files:
-        with bgzf.BgzfReader(input_file) as in_f:
-            shutil.copyfileobj(in_f, dest)
+
+    with gzip.open(dest, 'wt') as out_f:
+        for input_file in input_files:
+            with gzip.open(input_file, 'rt') as in_f:
+                shutil.copyfileobj(in_f, out_f)
 
 def merge_bam(bams: List[Path], dest: Path, threads: int) -> None:
 
@@ -172,19 +174,19 @@ def stitch_all_outputs(files: List[Path], output_dir) -> None:
             fq2_list.append(file)
         elif "r1.fastq" in file_name or ".fastq" in suffixes:
             fq1_list.append(file)
-        elif ".vcf" in suffixes:
+        elif ".vcf" in suffixes and ".tbi" not in suffixes:
             vcf_list.append(file)
-        elif ".bam" in suffixes:
+        elif ".bam" in suffixes and ".bai" not in suffixes:
             bam_list.append(file)
 
-    dest_fq1 = bgzf.BgzfWriter(f"{output_dir}/stitched_fq1.bgzf")
+    dest_fq1 = Path(f"{output_dir}/stitched_fq1.gz")
     dest_bam = Path(f"{output_dir}/stitched.bam")
     dest_vcf = Path(f"{output_dir}/stitched.vcf")
 
     concat_fq(fq1_list, dest_fq1)
 
     if (fq2_list):
-        dest_fq2 = bgzf.BgzfWriter(f"{output_dir}/stitched_fq2.bgzf")
+        dest_fq2 = Path(f"{output_dir}/stitched_fq2.gz")
         concat_fq(fq2_list, dest_fq2)
 
     merge_bam(bam_list, dest_bam, 2)

neat/read_simulator/utils/generate_reads.py

@@ -49,9 +49,9 @@ def cover_dataset(
     number_reads_per_layer = ceil(span_length / fragment_model.fragment_mean)
     if options.paired_ended:
         # TODO use gc bias to skew this number. Calculate at the runner level.
-        number_reads = number_reads_per_layer * (options.coverage//2)
+        number_reads = ceil(number_reads_per_layer * (options.coverage / 2))
     else:
-        number_reads = number_reads_per_layer * options.coverage
+        number_reads = ceil(number_reads_per_layer * options.coverage)
 
     # step 1: Divide the span up into segments drawn from the fragment pool. Assign reads based on that.
     # step 2: repeat above until number of reads exceeds number_reads

neat/read_simulator/utils/stitch_outputs.py

@@ -20,7 +20,7 @@
 def concat(files_to_join: List[Path], dest_file: gzip.GzipFile) -> None:
     if not files_to_join:
         # Nothing to do, and no error to throw
-        _LOG.warn(f"Concat called but there are no files to join: {files_to_join}" )
+        _LOG.warning(f"Concat called but there are no files to join: {files_to_join}" )
         return
     for f in files_to_join:
         with gzip.open(f, 'rt') as in_f:
@@ -37,7 +37,7 @@ def merge_vcfs(vcfs: List[Path], ofw: OutputFileWriter) -> None:
 def merge_bam(bam_files: List[Path], ofw: OutputFileWriter, threads: int):
     merged_file = ofw.tmp_dir / "temp_merged.bam"
     intermediate_files = []
-    # Note 1000 is abritrary. May need to be a user parameter/adjustable/a function
+    # Note 1000 is arbitrary. May need to be a user parameter/adjustable/a function
     for i in range(0, len(bam_files), 500):
         temp_file = str(ofw.tmp_dir / f"temp_merged_{i}.bam")
         pysam.merge("--no-PG", "-f", temp_file, *map(str, bam_files[i:i+500]))