replace output directory variable with pathvars

Author: ahmig

.gitignore

@@ -3,7 +3,7 @@
 
 # Analysis files
 /data/
-/output/
+/results/
 
 # MAC
 .DS_Store

.test/Snakefile

@@ -10,6 +10,9 @@ min_version("7.19")
 # Workflow version
 __version__ = "test"
 
+pathvars:
+    dataset = config["OUTPUT_NAME"]
+
 # Rules
 include: "../workflow/core.smk"
 include: "../workflow/rules/demix.smk"
@@ -19,4 +22,4 @@ include: "../workflow/rules/report.smk"
 
 rule all:
     input:
-        OUTDIR/f"{OUTPUT_NAME}.report.html"
+        "<results>/<dataset>/report.html"

.test/targets.yaml

@@ -10,8 +10,6 @@ SAMPLES:
     fasta: ".test/data/fasta/sample3.fasta"
 METADATA:
   ".test/data/metadata.csv"
-OUTPUT_DIRECTORY:
-  ".test/output"
 CONTEXT_FASTA:
   ".test/data/context.fasta"
 OUTPUT_NAME:

build_targets.py

@@ -91,7 +91,6 @@ def find_file_with_extension(directory: Path, prefix: str, extensions: List[str]
     else:
         sys.exit(f"ERROR: metadata file '{args.metadata_csv}' does not exist")
     targets["OUTPUT_NAME"] = args.output_name
-    targets["OUTPUT_DIRECTORY"] = "output"
     targets["CONTEXT_FASTA"] = None
     targets["MAPPING_REFERENCES_FASTA"] = None
 

config/README.md

@@ -37,8 +37,7 @@ The path to these input files is set in two configuration files in YAML format:
 [config.yaml](/config/config.yaml) (for general workflow settings) and
 [targets.yaml](/config/targets.yaml) (for specific dataset-related settings).
 The latter must be modified by the user to point the `SAMPLES` and `METADATA`
-parameters to your data. The `OUTPUT_DIRECTORY` parameter should point to your
-desired results directory.
+parameters to your data.
 
 The script [`build_targets.py`](/build_targets.py) simplifies the process of creating
 the targets configuration file. To run this script, you need to have PyYAML installed. It
@@ -62,8 +61,6 @@ SAMPLES:
   ...
 METADATA:
   "path/to/metadata.csv"
-OUTPUT_DIRECTORY:
-  "output"
 CONTEXT_FASTA:
   null
 MAPPING_REFERENCES_FASTA:

config/targets.yaml

@@ -39,9 +39,7 @@ METADATA:
   "data/metadata.csv"
 OUTPUT_NAME:
   "case_study"
-OUTPUT_DIRECTORY:
-  "output"
 CONTEXT_FASTA:
-  null
+  data/context.fasta
 MAPPING_REFERENCES_FASTA:
   null

workflow/Snakefile

@@ -14,6 +14,9 @@ containerized: "docker://ahmig/vipera:v" + __version__
 configfile: "config/config.yaml"
 configfile: "config/targets.yaml"
 
+pathvars:
+    dataset = config["OUTPUT_NAME"]
+
 include: "core.smk"
 include: "rules/context.smk"
 include: "rules/demix.smk"
@@ -22,4 +25,4 @@ include: "rules/report.smk"
 
 rule all:
     input:
-        OUTDIR/f"{OUTPUT_NAME}.report.html"
+        "<results>/<dataset>/report.html"

workflow/core.smk

@@ -2,16 +2,12 @@ BASE_PATH = Path(workflow.basedir).parent.resolve()
 
 include: "rules/common.smk"
 
-# Outputs
-OUTPUT_NAME = config["OUTPUT_NAME"]
-OUTDIR = Path(config["OUTPUT_DIRECTORY"])
-
 # Logging
-LOGDIR = OUTDIR / "logs"
+LOGDIR = Path("<results>/<logs>")
 
 # Report
-REPORT_DIR_PLOTS = Path(OUTDIR/"report/plots")
-REPORT_DIR_TABLES = Path(OUTDIR/"report/tables")
+REPORT_DIR_PLOTS = Path("<results>/<dataset>/report/plots")
+REPORT_DIR_TABLES = Path("<results>/<dataset>/report/tables")
 
 include: "rules/fetch.smk"
 include: "rules/fasta.smk"

workflow/rules/asr.smk

@@ -4,22 +4,21 @@ rule reconstruct_ancestral_sequence:
     params:
         seed = 7291,
         seqtype = "DNA",
-        name = OUTPUT_NAME,
         outgroup = config["ALIGNMENT_REFERENCE"],
         model = config["TREE_MODEL"]
     input:
-        fasta = OUTDIR/"nextalign"/f"{OUTPUT_NAME}.aligned.masked.fasta"
+        fasta = "<results>/<dataset>/aligned.masked.fasta"
     output:
-        folder = directory(OUTDIR/"tree"),
-        state_file = OUTDIR/"tree"/f"{OUTPUT_NAME}.state"
+        folder = directory("<results>/<dataset>/tree"),
+        state_file = "<results>/<dataset>/tree/asr.state"
     log:
         LOGDIR / "reconstruct_ancestral_sequence" / "log.txt"
     shell:
         "mkdir -p {output.folder} && "
         "iqtree2 -seed {params.seed} "
-            "-asr "
-            "-o {params.outgroup} -T AUTO --threads-max {threads} -s {input.fasta} "
-            "--seqtype {params.seqtype} -m {params.model} --prefix {output.folder}/{params.name} >{log} 2>&1"
+        "-asr "
+        "-o {params.outgroup} -T AUTO --threads-max {threads} -s {input.fasta} "
+        "--seqtype {params.seqtype} -m {params.model} --prefix {output.folder}/asr >{log} 2>&1"
 
 
 rule ancestor_fasta:
@@ -30,9 +29,9 @@ rule ancestor_fasta:
         indeterminate_char = "N",
         name = "case_ancestor",
     input:
-        state_file = OUTDIR/"tree"/f"{OUTPUT_NAME}.state"
+        state_file = "<results>/<dataset>/tree/asr.state"
     output:
-        fasta = report(OUTDIR/f"{OUTPUT_NAME}.ancestor.fasta")
+        fasta = report("<results>/<dataset>/ancestor.fasta")
     log:
         LOGDIR / "ancestor_fasta" / "log.txt"
     script:

workflow/rules/common.smk

@@ -49,7 +49,7 @@ def get_repo_version(base_dir: str, default: str, warn=False) -> str:
 def select_context_fasta():
     """Set context to be fetched automatically if CONTEXT_FASTA=null"""
     if "CONTEXT_FASTA" not in config.keys() or config["CONTEXT_FASTA"] is None:
-        return OUTDIR/"context"/"sequences.fasta"
+        return "<results>/<dataset>/context/sequences.fasta"
         if not Path(config["GISAID"]["CREDENTIALS"]).is_file():
             raise FileNotFoundError(f"Tried to download a context dataset, but no GISAID credentials were found at '{config['GISAID']['CREDENTIALS']}' (see README.md).")
     elif Path(config["CONTEXT_FASTA"]).is_file():
@@ -61,7 +61,7 @@ def select_context_fasta():
 def select_mapping_references_fasta():
     """Set mapping references to be fetched automatically if MAPPING_REFERENCES_FASTA=null"""
     if "MAPPING_REFERENCES_FASTA" not in config.keys() or config["MAPPING_REFERENCES_FASTA"] is None:
-        return OUTDIR/"mapping_references.fasta"
+        return "<results>/<dataset>/mapping_references.fasta"
     elif Path(config["MAPPING_REFERENCES_FASTA"]).is_file():
         return config["MAPPING_REFERENCES_FASTA"]
     else:
@@ -74,7 +74,7 @@ def is_url(string: str) -> bool:
 
 def select_problematic_vcf():
     if is_url(config["PROBLEMATIC_VCF"]):
-        return OUTDIR/"problematic_sites.vcf"
+        return "<results>/<dataset>/problematic_sites.vcf"
     elif Path(config["PROBLEMATIC_VCF"]).is_file():
         return config["PROBLEMATIC_VCF"]
     else: