Allow to extract dates for list of samples

Author: alexhbnr

amdirt/cli.py

@@ -169,6 +169,11 @@ def viewer(ctx, no_args_is_help=True, **kwargs):
     is_flag=True,
     help="Generate BibTeX file of all publications in input table",
 )
+@click.option(
+    "--dates",
+    is_flag=True,
+    help="Generate AncientMetagenomeDir dates table of all samples in input table"
+)
 @click.option(
     "--curl",
     is_flag=True,

amdirt/convert/__init__.py

@@ -8,6 +8,7 @@
     prepare_aMeta_table,
     is_merge_size_zero,
     prepare_taxprofiler_table,
+    get_dates,
     get_libraries,
     get_remote_resources,
     get_json_path,
@@ -27,6 +28,7 @@ def run_convert(
     tables=None,
     output=".",
     bibliography=False,
+    dates=False,
     librarymetadata=False,
     curl=False,
     aspera=False,
@@ -138,6 +140,22 @@ def run_convert(
     else:
         col_drop = ["archive_accession", "sample_host"]
 
+    if dates == True:
+        if table_name not in remote_resources["dates"]:
+            logger.error(f"No dates for {table_name} available in AncientMetagenomeDir at the moment.")
+        else:
+            tbl_file = f"{output}/AncientMetagenomeDir_filtered_dates.tsv"
+            dates_tbl = pd.read_csv(remote_resources['dates'][table_name],
+                                    sep="\t")
+            logger.info(f"Writing filtered dates table to {tbl_file}")
+            datesmetadata = get_dates(table_name, samples, dates_tbl)
+            datesmetadata.to_csv(
+                tbl_file,
+                sep="\t",
+                index=False,
+            )
+
+
     if librarymetadata == True:
         tbl_file = f"{output}/AncientMetagenomeDir_filtered_libraries.tsv"
         logger.info(f"Writing filtered libraries table to {tbl_file}")

amdirt/core/__init__.py

@@ -141,6 +141,40 @@ def doi2bib(doi: str) -> str:
     return r.text
 
 
+def get_dates(
+    table_name: str,
+    samples: pd.DataFrame,
+    dates: pd.DataFrame
+):
+    """Get dates from the samples table
+
+    Args:
+        table_name (str): Name of the table of the table to convert
+        samples (pd.DataFrame): Sample table
+
+    Returns:
+        pd.DataFrame: filtered dates table
+    """
+    samples = (samples
+        .rename({'archive_accession': 'archive_sample_accession'}, axis=1)
+    )
+    if table_name in [
+        "ancientmetagenome-environmental",
+    ]:
+        sel_col = ["archive_accession"]
+    else:
+        sel_col = ["project_name", "publication_year",
+                   "sample_name", "singlegenome_species",
+                   "archive_project", "archive_sample_accession"]
+    selected_dates = dates.merge(
+        samples.loc[:, sel_col],
+        on=sel_col,
+        how="inner"
+    )
+
+    return selected_dates
+
+
 @st.cache_data
 def get_libraries(
     table_name: str,