feat: update bigsky module initialization

.gitignore

@@ -1,6 +1,7 @@
 logs
 .snakemake
 site
+__pycache__
 output
 .tests/illumnia_demux/dry_run_out
 .tests/dry_run_out

scripts/config.py

@@ -23,7 +23,7 @@ def get_current_server():
     # biowulf hostnames
     re_biowulf_head = (r"biowulf\.nih\.gov", "biowulf")
     re_biowulf_compute = (r"cn\d{4}", "biowulf")
-    
+
     # skyline hostnames
     re_skyline_head = (r"ai-hpc(submit|n)(\d+)?", "skyline")
     re_skyline_compute = (r"ai-hpc(submit|n)(\d+)?", "skyline")
@@ -48,15 +48,15 @@ def get_current_server():
 FRCE_PATH = "COVID-19_Consortium"
 
 
-# ~~~ labkey configurations ~~~ 
+# ~~~ labkey configurations ~~~
 CONTEXT_PATH = "labkey"
 LABKEY_CONFIGS = {
     "bigsky": {"domain": BIGSKY_DEV, "container_path": BIGSKY_PATH, "context_path": CONTEXT_PATH, "use_ssl": True},
     "frce": {"domain": FRCE_PROD, "container_path": FRCE_PATH, "context_path": CONTEXT_PATH, "use_ssl": True}
 }
 
 
-# ~~~ snakemake configurations ~~~ 
+# ~~~ snakemake configurations ~~~
 illumina_pipelines = defaultdict(lambda: Path(Path(__file__).parent.parent, "workflow", "Snakefile").resolve())
 # can add support for NextSeq2k and bclconvert here
 SNAKEFILE = {
@@ -69,7 +69,7 @@ def get_current_server():
 
 
 def get_resource_config():
-    """Return a dictionary containing server specific references utilized in 
+    """Return a dictionary containing server specific references utilized in
     the workflow for directories or reference files.
 
     Returns:
@@ -141,7 +141,7 @@ def get_bigsky_seq_dirs():
         for this_child_elem in this_dir.iterdir():
             try:
                 elem_checks = [
-                    this_child_elem.is_dir(), 
+                    this_child_elem.is_dir(),
                     Path(this_child_elem, transfer_breadcrumb).exists(),
                     check_access(this_child_elem, R_OK)
                 ]
@@ -155,13 +155,13 @@ def get_bigsky_seq_dirs():
 def get_tmp_dir(host):
     TMP_CONFIGS = {
         'skyline': {'user': '/data/scratch/$USER/$SLURM_JOBID', 'global': '/data/scratch/$USER/' + str(uuid4())},
-        'bigsky': {'user': '/gs1/Scratch/$USER/$SLURM_JOBID', 'global': '/gs1/Scratch/$USER/' + str(uuid4())},
+        'bigsky': {'user': '/data/scratch/$USER/$SLURM_JOBID', 'global': '/data/scratch/$USER/' + str(uuid4())},
         'biowulf': {'user': '/lscratch/$SLURM_JOBID', 'global': '/tmp/$USER/' + str(uuid4())}
     }
 
     this_tmp = TMP_CONFIGS[host]['user']
 
-    # this directory, if it does not exist, 
+    # this directory, if it does not exist,
     if Path(this_tmp).parents[0].exists():
         return this_tmp
     else:
@@ -222,4 +222,4 @@ def get_tmp_dir(host):
         "mesaur": "/data/openomics/references/genomes/mesaur/2.0/GCF_017639785.1_BCM_Maur_2.0_genomic.fna.gz",
         "cynomac": "/data/openomics/references/genomes/cynomac/v2/GCF_012559485.2_MFA1912RKSv2_genomic.fna.gz",
     },
-}
+}

scripts/files.py

@@ -13,7 +13,7 @@
 
 def get_all_seq_dirs(top_dir, server):
     """
-        Gather and return all sequencing directories from the `top_dir`. 
+        Gather and return all sequencing directories from the `top_dir`.
         This is tightly coupled at the moment to the directory that is on RML-BigSky.
         In the future will need to the take a look at how to do this more generally
     """
@@ -42,7 +42,7 @@ def valid_run_output(output_directory, dry_run=False):
     output_directory = Path(output_directory).absolute()
     if not output_directory.exists():
         output_directory.mkdir(parents=True, mode=0o765)
-    
+
     if not check_access(output_directory, W_OK):
         raise PermissionError(f'Can not write to output directory {output_directory}')
     return output_directory
@@ -70,7 +70,7 @@ def valid_fasta(suspect):
 
     if not is_valid:
         raise ValueError
-            
+
     return suspect
 
 
@@ -147,7 +147,7 @@ def find_demux_dir(run_dir):
 
     if len(demux_stat_files) != 1:
         raise FileNotFoundError
-    
+
     return Path(demux_stat_files[0], '..').absolute()
 
 
@@ -160,7 +160,7 @@ def get_run_directories(runids, seq_dir=None, sheetname=None):
             for secondchild in firstchild.iterdir():
                 seq_contents.append(secondchild)
     seq_contents_names = [child for child in map(lambda d: d.name, seq_contents)]
-    
+
     run_paths, invalid_runs  = [], []
     run_return = []
     for run in runids:
@@ -192,7 +192,7 @@ def get_run_directories(runids, seq_dir=None, sheetname=None):
             sheet = Path(run_p, sheetname).absolute()
         else:
             raise FileNotFoundError(f'Run {rid}({run_p}) does not have a find-able sample sheet.')
-        
+
         this_run_info['samplesheet'] = parse_samplesheet(sheet)
         this_run_info.update({info.tag: info.text for run in runinfo_xml.getroot() for info in run \
                              if info.text is not None and info.text.strip() not in ('\n', '')})
@@ -201,5 +201,5 @@ def get_run_directories(runids, seq_dir=None, sheetname=None):
     if invalid_runs:
         raise ValueError('Runs entered are invalid (missing sequencing artifacts or directory does not exist): \n' + \
                          ', '.join(invalid_runs))
-    
-    return run_return
+
+    return run_return

scripts/utils.py

@@ -97,7 +97,7 @@ def valid_run_input(run):
 
 
 def exec_snakemake(popen_cmd, local=False, dry_run=False, env=None, cwd=None):
-    # async execution w/ filter: 
+    # async execution w/ filter:
     #   - https://gist.github.com/DGrady/b713db14a27be0e4e8b2ffc351051c7c
     #   - https://lysator.liu.se/~bellman/download/asyncproc.py
     #   - https://gist.github.com/kalebo/1e085ee36de45ffded7e5d9f857265d0
@@ -113,7 +113,6 @@ def exec_snakemake(popen_cmd, local=False, dry_run=False, env=None, cwd=None):
         popen_kwargs['cwd'] = cwd
     else:
         popen_kwargs['cwd'] = str(Path.cwd())
-
     parent_jobid = None
     if local or dry_run:
         popen_kwargs['env'].update(os.environ)
@@ -170,9 +169,7 @@ def get_mods(init=False):
     mod_cmd = []
 
     if host == 'bigsky':
-        mod_cmd.append('source /gs1/apps/user/rmlspack/share/spack/setup-env.sh')
-        mod_cmd.append('spack load miniconda3@4.11.0')
-        mod_cmd.append('source activate snakemake7-19-1')
+        mod_cmd.append('module load snakemake/7.22.0-ufanewz')
     elif host == 'skyline':
         mod_cmd.append('source /data/openomics/bin/dependencies.sh')
     elif host == 'biowulf':
@@ -220,8 +217,9 @@ def get_mounts(*extras):
                 raise FileNotFoundError(f"Can't mount {str(bind)}, it doesn't exist!")
             file_to, file_from, mode = str(bind), str(bind), 'rw'
         mounts.append(file_from + ':' + file_to + ':' + mode)
-
-    mounts.append(r'\$TMPDIR:/tmp:rw')
+
+    if 'TMPDIR' in os.environ:
+        mounts.append(os.environ['TMPDIR'] + ':/tmp:rw')
 
     return ','.join(mounts)
 
@@ -265,13 +263,15 @@ def exec_pipeline(configs, dry_run=False, local=False):
         top_env['PATH'] = os.environ["PATH"]
         top_env['SNK_CONFIG'] = str(config_file.absolute())
         top_env['SINGULARITY_CACHEDIR'] = str(Path(this_config['out_to'], '.singularity').absolute())
+        top_env['SINGULARITY_CONTAINALL'] = '1'
+        top_env['APPTAINER_CONTAINALL'] = '1'
         this_cmd = [
-            "snakemake", "-p", "--use-singularity", "--rerun-incomplete", "--keep-incomplete",
-            "--rerun-triggers", "mtime", "--verbose", "-s", snake_file,
+            "snakemake", "-p", "--cores", "2", "--use-singularity", "--rerun-incomplete", "--keep-incomplete",
+            "--rerun-triggers", "mtime", "--verbose", "-s", str(snake_file),
         ]
 
         if singularity_binds and not dry_run:
-            this_cmd.extend(["--singularity-args", f"\"--env 'TMPDIR=/tmp' -C -B '{singularity_binds}'\""])
+            this_cmd.extend(["--singularity-args", f"\"-B '{singularity_binds}'\""])
 
         if dry_run:
             print(f"{esc_colors.OKGREEN}> {esc_colors.ENDC}{esc_colors.UNDERLINE}Dry run{esc_colors.ENDC} " + \
@@ -314,4 +314,4 @@ def valid_host_pathogen_genomes(host, pathogen):
         if not g2:
             raise ValueError('Pathogen genome does not exist on the file system.')
 
-    return host, pathogen
+    return host, pathogen

workflow/fastq.smk

@@ -76,7 +76,7 @@ rule kaiju_annotation:
     log: config['out_to'] + "/logs/" + config["project"] + "/kaiju/{sids}.log",
     threads: 24
     resources: 
-        mem_mb = 220000, 
+        mem_mb = 300000, 
         runtime = 60*24*2
     shell:
         """
@@ -109,7 +109,7 @@ rule kraken_annotation:
     log: config['out_to'] + "/logs/" + config["project"] + "/kraken/{sids}.log",
     threads: 24
     resources: 
-        mem_mb = 220000,
+        mem_mb = 300000,
         runtime = 60*24*2
     shell:
         """