mgnify-pipelines-toolkit 1.2.7__tar.gz → 1.2.9__tar.gz

{mgnify_pipelines_toolkit-1.2.7 → mgnify_pipelines_toolkit-1.2.9}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mgnify_pipelines_toolkit
-Version: 1.2.7
+Version: 1.2.9
 Summary: Collection of scripts and tools for MGnify pipelines
 Author-email: MGnify team <metagenomics-help@ebi.ac.uk>
 License: Apache Software License 2.0

{mgnify_pipelines_toolkit-1.2.7 → mgnify_pipelines_toolkit-1.2.9}/mgnify_pipelines_toolkit/analysis/amplicon/make_asv_count_table.py

@@ -300,7 +300,7 @@ def main():
     if paired_end:
         rev_fr.close()
 
-    if asv_dict:   # if there are matches between taxonomic and ASV annotations
+    if asv_dict:  # if there are matches between taxonomic and ASV annotations
         ref_db = ""
 
         if len(taxa_df.columns) == 9:

{mgnify_pipelines_toolkit-1.2.7 → mgnify_pipelines_toolkit-1.2.9}/mgnify_pipelines_toolkit/analysis/assembly/antismash_gff_builder.py

@@ -89,8 +89,10 @@ def main():
             if iter_cds and feature["type"] == "CDS":
                 # Annotate CDS features
 
-                start = int(feature["location"].split(":")[0][1:])
-                end = int(feature["location"].split(":")[1].split("]")[0])
+                # The > and < are removed to work with pseudogene outputs in Bakta
+                # A feature["location"] example that can be seen in Bakta outputs: "[81883:>82231](+)"
+                start = int(feature["location"].split(":")[0][1:].lstrip("<>"))
+                end = int(feature["location"].split(":")[1].split("]")[0].lstrip("<>"))
                 strand = feature["location"].split("(")[1][0]  # + or -
 
                 if not region_name or not (region_start <= end and start <= region_end):

{mgnify_pipelines_toolkit-1.2.7 → mgnify_pipelines_toolkit-1.2.9}/mgnify_pipelines_toolkit/analysis/assembly/summarise_sanntis_bgcs.py

@@ -110,10 +110,9 @@ def main():
         df_merged = df_merged[
             ["nearest_mibig", "nearest_mibig_class", "description", "count"]
         ]
-        df_merged = df_merged.rename(columns={
-            "Description": "description",
-            "Count": "count"
-        })
+        df_merged = df_merged.rename(
+            columns={"Description": "description", "Count": "count"}
+        )
         df_merged.to_csv(output_filename, sep="\t", index=False)
 
 

{mgnify_pipelines_toolkit-1.2.7 → mgnify_pipelines_toolkit-1.2.9}/mgnify_pipelines_toolkit/analysis/rawreads/study_summary_generator.py

@@ -53,7 +53,7 @@ def cli():
 
 def get_file(
     run_acc: str, analyses_dir: Path, db_label: str
-) -> Union[Path, List[Path]]:
+) -> Union[Path, List[Path], None]:
     """Takes path information for a particular analysis and db_label combo, and returns any existing files.
 
     :param run_acc: Run accession for the tax file that should be retrieved.
@@ -84,7 +84,7 @@ def get_file(
         return
 
     analysis_file = Path(
-        f"{analyses_dir}/{run_acc}/{db_dir}/{db_label}/{run_acc}_{db_label}.txt"
+        f"{analyses_dir}/{run_acc}/{db_dir}/{db_label}/{run_acc}_{db_label}.txt.gz"
     )
     if not analysis_file.exists():
         logging.error(
@@ -119,20 +119,25 @@ def parse_one_tax_file(run_acc: str, tax_file: Path, db_label: str) -> pd.DataFr
     :rtype: pd.DataFrame
     """
 
-    tax_ranks = _MOTUS_TAX_RANKS if db_label == "mOTUs" else _SILVA_TAX_RANKS
+    tax_ranks = _MOTUS_TAX_RANKS if db_label == "motus" else _SILVA_TAX_RANKS
     res_df = pd.read_csv(tax_file, sep="\t", skiprows=1, names=["Count"] + tax_ranks)
     res_df = res_df.fillna("")
 
-    validate_dataframe(
-        res_df, MotusTaxonSchema if db_label == "mOTUs" else TaxonSchema, str(tax_file)
-    )
+    if res_df.shape[0] > 0:
+        validate_dataframe(
+            res_df,
+            MotusTaxonSchema if db_label == "motus" else TaxonSchema,
+            str(tax_file),
+        )
 
-    res_df["full_taxon"] = res_df.iloc[:, 1:].apply(
-        lambda x: ";".join(x).strip(";"), axis=1
+    res_df["full_taxon"] = [
+        ";".join(r[tax_ranks]).strip(";") for _, r in res_df.iterrows()
+    ]
+    final_df = (
+        res_df[["Count", "full_taxon"]]
+        .set_index("full_taxon")
+        .rename(columns={"Count": run_acc})
     )
-    final_df = res_df.iloc[:, [0, -1]]
-    final_df = final_df.set_index("full_taxon")
-    final_df.columns = [run_acc]
 
     return final_df
 
@@ -162,16 +167,20 @@ def parse_one_func_file(
     ).set_index("function")
     res_df = res_df.fillna(0)
 
-    validate_dataframe(res_df, FunctionProfileSchema, str(func_file))
+    if res_df.shape[0] > 0:
+        validate_dataframe(res_df, FunctionProfileSchema, str(func_file))
 
-    count_df = res_df[["read_count"]]
-    count_df.columns = [run_acc]
+    count_df = pd.DataFrame(res_df[["read_count"]]).rename(
+        columns={"read_count": run_acc}
+    )
 
-    depth_df = res_df[["coverage_depth"]]
-    depth_df.columns = [run_acc]
+    depth_df = pd.DataFrame(res_df[["coverage_depth"]]).rename(
+        columns={"coverage_depth": run_acc}
+    )
 
-    breadth_df = res_df[["coverage_breadth"]]
-    breadth_df.columns = [run_acc]
+    breadth_df = pd.DataFrame(res_df[["coverage_breadth"]]).rename(
+        columns={"coverage_breadth": run_acc}
+    )
 
     return count_df, depth_df, breadth_df
 
@@ -423,7 +432,9 @@ def merge_summaries(analyses_dir: str, output_prefix: str) -> None:
                         curr_df = pd.read_csv(summary, sep="\t", index_col=0)
                         res_df = res_df.join(curr_df, how="outer")
                         res_df = res_df.fillna(0)
-                        res_df = res_df.astype(int if table_type == "count" else float)
+                        res_df = res_df.astype(
+                            int if table_type == "read-count" else float
+                        )
 
                     res_df = res_df.reindex(sorted(res_df.columns), axis=1)
                     res_df.to_csv(