small changes from issues that came up in real analysis scenarios

clemente-lab · Jan 15, 2025 · c2a6317 · c2a6317
1 parent 5125c48
commit c2a6317
Show file tree

Hide file tree

Showing 4 changed files with 6 additions and 4 deletions.
diff --git a/mmeds/snakemake/rules/common.smk b/mmeds/snakemake/rules/common.smk
@@ -11,7 +11,7 @@ def pairwise_splits(wildcards, tool, vars):
     else:
         tables = [f"taxa_table_L{x}" for x in config["taxa_levels"]]
 
-    subclasses = None
+    subclasses = False
     if tool == "lefse" and "subclasses" in config and config["subclasses"]:
         subclasses = deepcopy(config["subclasses"])
 

diff --git a/mmeds/snakemake/rules/demux_denoise.smk b/mmeds/snakemake/rules/demux_denoise.smk
@@ -31,7 +31,7 @@ rule demux_dual_barcodes_pheniqs:
 
 rule strip_error_barcodes:
     input:
-        dir = directory("section_{sequencing_run}/pheniqs_output"),
+        dir = "section_{sequencing_run}/pheniqs_output",
         mapping_file = "section_{sequencing_run}/qiime_mapping_file_{sequencing_run}.tsv",
     output:
         dir = directory("section_{sequencing_run}/stripped_output")

diff --git a/mmeds/snakemake/rules/file_manipulation.smk b/mmeds/snakemake/rules/file_manipulation.smk
@@ -41,15 +41,15 @@ rule format_metadata_qiime_to_lefse_class:
     output:
         "tables/{class}/lefse_format.{table}.{class}.{subclass}.tsv"
     params:
-        subclass = lefse_get_subclass
+        subclass_param = lefse_get_subclass
     conda:
         "mmeds_test"
     shell:
         "format_lefse.py "
         "-i {input.feature_table} "
         "-m {input.mapping_file} "
         "-c {wildcards.class} "
-        "-s {params.subclass} "
+        "-s {params.subclass_param} "
         "-u HostSubjectId "
         "-o {output}"
 

diff --git a/mmeds/util.py b/mmeds/util.py
@@ -1704,6 +1704,7 @@ def format_table_to_lefse(i_table, metadata_file, metadata_column_class, metadat
     # Store metadata by ID, allowing for any subset of samples from the metadata
     categories = {}
     for i, cell in enumerate(mdf['#SampleID']['#q2:types']):
+        cell = str(cell)
         if cell not in categories:
             categories[cell] = {}
         categories[cell][metadata_column_class] = mdf[metadata_column_class]['categorical'][i]
@@ -1725,6 +1726,7 @@ def format_table_to_lefse(i_table, metadata_file, metadata_column_class, metadat
     t = [metadata_column_class]
     to_drop = []
     for i, cell in enumerate(path_df.loc[0]):
+        cell = str(cell)
         if i == 0:
             continue
         if pd.isna(categories[cell][metadata_column_class]):