feat: pass Snakemake profile to ZARP workflow (#77)

docs/guides/initialization.md

@@ -45,6 +45,7 @@ The following configuration options are available.
 | `genome_assemblies_map` | A headerless 3-column semicolon-separated mapping table of organism/source trivial names (e.g., `homo_sapiens`), optional comma-separated aliases such as NCBI taxon IDs and/or organism/source short names (e.g., `7227,dmelanogaster`) and a corresponding genome assembly name (e.g., `GRCm39`); a table in the required format is shipped with _ZARP_cli_ in the location provided in the default location; which can be amended with additional aliases; note that for [`genomepy`][genomepy] to be able to pull genome annotations for organisms/sources that [HTSinfer][htsinfer] inferred, NCBI taxon ID aliases are _required_  | `./data/genome_assemblies.map` relative to the location of the ZARP-cli repository |
 | `resources_version` | Whether to always download the latest available version of genome annotations for a given organism/source from Ensembl (enter `None`; default) or whether to use a specific version of the corresponding Ensembl database (e.g., `100`); note that the different Ensembl databases (e.g., for fungi, plants) use a different versioning scheme, so pinning a particular database version may lead to unexpected outcomes | `None` |
 | `rule_config` | A configuration file for the _ZARP_ workflow that sets specific parameters for each workflow step ("rule"); see [ZARP][zarp] documentation for details | `None` |
+| `profile` | Path to [Snakemake profile][snakemake-profiles] to be used for the _ZARP_ workflow. Use this to optimize _ZARP_ for your specific compute environment |
 | `fragment_length_distribution_mean` | HTSinfer currently is unable to infer the mean of the fragment length distribution of RNA-seq libraries; however, this value is required for tools [`kallisto`][kallisto] and [`salmon`][salmon] -which are executed as part of _ZARP_- when run on single-ended libraries only (for paired-ended libraries, the tools are able to infer this parameter from the data); the value provided here is used as a fallback if the value was not determined experimentally (e.g., with [Bioanalyzer][bioanalyzer] instruments) and provided via a sample table | `300` |
 | `fragment_length_distribution_sd` | Analogous to `fragment_length_distribution_mean` above, but this parameter is for the _standard deviation_ of the fragment length distribution | `100` |
 | `author` | Name of the person or organization executing the _ZARP-cli_ runs; will be added to the _ZARP_ report | `None` |

docs/includes/references.md

@@ -11,6 +11,7 @@
 [salmon]: <https://github.com/COMBINE-lab/salmon>
 [singularity]: <https://sylabs.io/singularity/>
 [snakemake]: <https://github.com/snakemake/snakemake>
+[snakemake-profiles]: <https://snakemake.readthedocs.io/en/stable/executing/cli.html#profiles>
 [sra]: <https://www.ncbi.nlm.nih.gov/sra>
 [sra-toolkit]: <https://github.com/ncbi/sra-tools>
 [zarp]: <https://github.com/zavolanlab/zarp>

tests/snakemake/test_run.py

@@ -73,6 +73,20 @@ def test_compile_command_config_file(self, tmpdir):
         cmd = my_run.compile_command(snakefile=snakefile)
         assert "--configfile" in cmd
 
+    def test_compile_command_profile(self, tmpdir):
+        """Execute a run with a Snakemake profile."""
+        os.chdir(tmpdir)
+        snakefile = create_snakefile(dir=Path(tmpdir))
+        run_config = default_run_config.copy(deep=True)
+        run_config.profile = Path("/path/to/profile")
+        my_run = SnakemakeExecutor(
+            run_config=run_config,
+            exec_dir=tmpdir,
+        )
+        cmd = my_run.compile_command(snakefile=snakefile)
+        assert "--profile" in cmd
+        os.chdir(default_cwd)
+
     @pytest.mark.parametrize(
         "dependency_embedding",
         [

zarp/config/args.py

@@ -41,6 +41,7 @@ class ArgParser:
             "execution_mode",
             "genome_assemblies_map",
             "identifier",
+            "profile",
             "resources_version",
             "rule_config",
             "working_directory",
@@ -425,6 +426,16 @@ def _set_run_arguments(
                 " generated"
             ),
         )
+        argument_group.add_argument(
+            "--profile",
+            default=None,
+            type=lambda p: Path(p).absolute(),
+            metavar="PATH",
+            help=(
+                "Snakemake profile for ZARP workflow; refer to ZARP"
+                " documentation for details"
+            ),
+        )
         argument_group.add_argument(
             "--resources-version",
             default=None,

zarp/config/models.py

@@ -73,6 +73,7 @@ class InitRun(CustomBaseModel):
         dependency_embedding: Dependency embedding strategy to use.
         execution_mode: Execution mode to use.
         genome_assemblies_map: Genome assemblies mapping file.
+        profile: Snakemake profile for ZARP workflow.
         resources_version: Version of Ensembl genome resources to use when
             resources are not provided.
         rule_config: ZARP rule configuration.
@@ -84,6 +85,7 @@ class InitRun(CustomBaseModel):
         dependency_embedding: Dependency embedding strategy to use.
         execution_mode: Execution mode to use.
         genome_assemblies_map: Genome assemblies mapping file.
+        profile: Snakemake profile for ZARP workflow.
         resources_version: Version of Ensembl genome resources to use when
             resources are not provided.
         rule_config: ZARP rule configuration.
@@ -103,6 +105,7 @@ class InitRun(CustomBaseModel):
     )
     resources_version: Optional[int] = None
     rule_config: Optional[Path] = None
+    profile: Optional[Path] = None
 
 
 class InitSample(CustomBaseModel):

zarp/snakemake/run.py

@@ -63,6 +63,9 @@ def compile_command(self, snakefile: Path) -> List[str]:
         cmd_ls.extend(["--directory", str(self.exec_dir)])
         if self.config_file is not None:
             cmd_ls.extend(["--configfile", str(self.config_file)])
+        if self.run_config.profile is not None:
+            if snakefile.name == "Snakefile":
+                cmd_ls.extend(["--profile", str(self.run_config.profile)])
         if self.run_config.execution_mode == "DRY_RUN":
             cmd_ls.append("--dry-run")
         if self.run_config.dependency_embedding == "CONDA":

zarp/version.py

@@ -1,3 +1,3 @@
 """Single source of truth for package version."""
 
-__version__ = "0.3.0"
+__version__ = "0.4.0"