update genomepy + gimmemotifs + fix install

requirements.yaml

@@ -5,9 +5,9 @@ channels:
 dependencies:
   - conda-forge::python=3.10
   - bioconda::snakemake-minimal=7.25.0
-  - conda-forge::conda=24.1.2
-  - conda-forge::mamba=1.5.9
-  - bioconda::genomepy=0.16.2
+  - conda-forge::conda=23.5.2
+  - conda-forge::mamba=1.5.1
+  - bioconda::genomepy=0.16.3
   - conda-forge::biopython=1.81
   - conda-forge::filelock=3.10.7
   - conda-forge::pyyaml=6.0
@@ -20,3 +20,5 @@ dependencies:
   - conda-forge::argcomplete=3.0.5
   - conda-forge::tabulate=0.9.0
   - conda-forge::conda-ecosystem-user-package-isolation=1.0
+  - conda-forge::setuptools=59.4.0
+  - conda-forge::toml=0.10.2

seq2science/envs/assembly_stats.yaml

@@ -5,5 +5,5 @@ channels:
 dependencies:
   - conda-forge::python=3.9
   - conda-forge::matplotlib-base=3.7.1
-  - bioconda::genomepy=0.15.0
+  - bioconda::genomepy=0.16.3
   - conda-forge::conda-ecosystem-user-package-isolation=1.0

seq2science/envs/gimme.yaml

@@ -3,7 +3,7 @@ channels:
   - conda-forge
   - bioconda
 dependencies:
-  - bioconda::gimmemotifs-minimal=0.18.0
+  - bioconda::gimmemotifs-minimal=0.18.1
   - bioconda::gffread=0.12.7
   - bioconda::orthofinder=2.5.4
   - conda-forge::xgboost=1.0.2

seq2science/envs/pytxi.yaml

@@ -3,7 +3,7 @@ channels:
   - conda-forge
   - bioconda
 dependencies:
-  - bioconda::genomepy=0.15.0
+  - bioconda::genomepy=0.16.3
   - conda-forge::conda-ecosystem-user-package-isolation=1.0
   - pip:
     - git+https://github.com/vanheeringen-lab/pytxi.git@v0.1.2

seq2science/rules/configuration_generic.smk

@@ -336,10 +336,10 @@ if WORKFLOW != "download_fastq":
         used_assemblies = list(set(samples["assembly"]))
         # we make a temporary _used assemblies as gimme maelstrom might need more assemblies downloaded
         # and those assemblies need to be saved in the local/remote assemblies variable
-        if "motif2factors_reference" in config and config["run_gimme_maelstrom"]:
+        _used_assemblies = used_assemblies
+        if (config.get("run_gimme_maelstrom") and config.get("infer_motif2factors") and
+                set(samples["assembly"]) - set(config.get("motif2factors_database_references"))):
             _used_assemblies = used_assemblies + config["motif2factors_reference"] + config["motif2factors_database_references"]
-        else:
-            _used_assemblies = used_assemblies
 
         # dictionary with which providers to use per genome
         providers = PickleDict(os.path.join(CACHE_DIR, "providers.p"))

seq2science/rules/configuration_logging.smk

@@ -154,6 +154,8 @@ if not config.get("no_config_log"):
         ("min_mapping_quality", no_aligners),
         ("only_primary_align", no_aligners),
         ("remove_blacklist", no_aligners),
+        ("motif2factors_database_references", not config.get("motif2factors_database_references")),
+        ("motif2factors_reference", not config.get("motif2factors_reference")),
         ("tx2gene_from_gtf", config.get("quantifier") != "salmon"),
         ("tximeta", config.get("quantifier") != "salmon"),
         ("deseq2", not config.get("contrasts")),

seq2science/rules/configuration_workflows.smk

@@ -71,6 +71,18 @@ if config.get("peak_caller", False):
             "To run gimme maelstrom you need more than one biological replicate!"
         )
 
+        # only need additional assemblies when ortholog inference is truly needed
+        if config["infer_motif2factors"] is True:
+            if len(set(samples["assembly"]) - set(config["motif2factors_database_references"])) == 0:
+                config["infer_motif2factors"] = False
+        if config["infer_motif2factors"] is False:
+            config["motif2factors_database_references"] = []
+            config["motif2factors_reference"] = []
+    else:
+        config["infer_motif2factors"] = False
+        config["motif2factors_database_references"] = []
+        config["motif2factors_reference"] = []
+
 # make sure that both maximum and minimum insert sizes are existing when one of them is used
 if config.get("min_template_length") and not config.get("max_template_length"):
     config["max_template_length"] = 1_000_000_000

seq2science/util.py

@@ -1039,12 +1039,21 @@ def _get_current_version(package):
     """
     Attempt to return a given package's version
     """
-    # package-isolation is not a package
+    # conda-ecosystem-user-package-isolation is not a package
     # xdg keeps its version in a pyproject.toml (not included)
     # argcomplete keeps its version in a setup.py (not included)
     # trackhub versioning is weird
     # mamba is not a package
-    if package in ["conda-ecosystem-user-package-isolation", "xdg", "argcomplete", "trackhub", "mamba"]:
+    # setuptools and toml are for packaging only
+    if package in [
+        "conda-ecosystem-user-package-isolation",
+        "xdg",
+        "argcomplete",
+        "trackhub",
+        "mamba",
+        "setuptools",
+        "toml",
+    ]:
         return None
     if package == "python":
         return sys.version.split()[0]