Added p value to Cor, Changes s.t. can run terms q

BackofenLab · Jul 3, 2023 · b525a5c · b525a5c
1 parent 87823d0
commit b525a5c
Show file tree

Hide file tree

Showing 4 changed files with 67 additions and 12 deletions.
diff --git a/db/scripts/read/read_experiment.py b/db/scripts/read/read_experiment.py
@@ -43,6 +43,11 @@ def read_experiment():
                     df=pd.read_csv(file, sep="\t"), file_name=file_name.split("/")[-1], reformat=reformat
                 )
                 dataframes[index] = df
+            elif file_extention == ".csv":
+                df, index = _reformat_experiment_file(
+                    df=pd.read_csv(file, sep=","), file_name=file_name.split("/")[-1], reformat=reformat
+                )
+                dataframes[index] = df
         return dataframes
 
     def filter_df_by_context(context: str, df: pd.DataFrame, protein: bool):
@@ -129,7 +134,7 @@ def post_processing(exp: list[pd.DataFrame]):
 
         # Filter for relevant columns
         tmp_or_id = relevant_info.filter(items=["id", "nearest_index"])
-        or_tg_corr = exp[3].filter(items=["ENSEMBL", "Correlation", "nearest_index"])
+        or_tg_corr = exp[3].filter(items=["ENSEMBL", "Correlation", "nearest_index", "p"])
         or_tg_corr = or_tg_corr.merge(tmp_or_id, left_on="nearest_index", right_on="nearest_index", how="left")
         or_tg_corr = or_tg_corr.drop(columns=["nearest_index"])
 
@@ -158,7 +163,7 @@ def _reformat_experiment_file(df: pd.DataFrame, file_name: str, reformat: bool):
     print_update(update_type="Reformatting", text=file_name, color="orange")
 
     # Filename and function pairs: same index <-> use function for file
-    names = ["exp_DA", "exp_DE_filter", "TF_target_cor_", "peak_target_cor_", "TF_motif_peak"]
+    names = ["exp_DA", "exp_DE_filter", "correlation_pval_TF_target", "corr_peak_target", "TF_motif_peak"]
     functions = [_reformat_da, _reformat_de, _reformat_tf_tg, _reformat_or_tg, _reformat_motif]
     index = names.index(file_name)
 
@@ -184,12 +189,14 @@ def _reformat_de(df: pd.DataFrame):
 
 
 def _reformat_tf_tg(df: pd.DataFrame):
-    df = df.rename(columns={"nearest_ENSEMBL": "ENSEMBL", "TF_target_cor": "Correlation"})
+    df = df.filter(items=["nearest_ENSEMBL_target", "ENSEMBL_TF", "korrelationskoeffizient", "p-Wert"])
+    df = df.rename(columns={"nearest_ENSEMBL_target": "ENSEMBL_TG", "korrelationskoeffizient": "Correlation", "p-Wert": "p"})
     return df
 
 
 def _reformat_or_tg(df: pd.DataFrame):
-    df = df.rename(columns={"nearest_ENSEMBL": "ENSEMBL", "peak_target_cor": "Correlation"})
+    df = df.filter(items=["nearest_ENSEMBL_x", "nearest_index", "correlation_coefficient", "p-value"])
+    df = df.rename(columns={"nearest_ENSEMBL_x": "ENSEMBL", "correlation_coefficient": "Correlation", "p-value": "p"})
     return df
 
 

diff --git a/db/scripts/upload/upload_base.py b/db/scripts/upload/upload_base.py
@@ -134,10 +134,11 @@ def create_functional(
     """
     print_update(update_type="Node Creation", text="Functional Term", color="blue")
 
+    # TODO: Change "Terms" to "FT"
     save_df_to_csv(file_name="ft_nodes.csv", df=ft_nodes, override_prod=True)
     create_nodes(
         source_file="ft_nodes.csv",
-        type_="FT",
+        type_="Terms",
         id="Term",
         values=["Term", "Name", "Category"],
         reformat_values=[],
@@ -152,7 +153,7 @@ def create_functional(
         source_file="ft_overlap.csv",
         type_="OVERLAP",
         between=(("Term", "source"), ("Term", "target")),
-        node_types=("FT", "FT"),
+        node_types=("Terms", "Terms"),
         values=["Score"],
         reformat_values=[("Score", "toFloat")],
         merge=False,
@@ -166,7 +167,7 @@ def create_functional(
         source_file="ft_gene.csv",
         type_="LINK",
         between=(("ENSEMBL", "ENSEMBL"), ("Term", "Term")),
-        node_types=("TG", "FT"),
+        node_types=("TG", "Terms"),
         values=[],
         reformat_values=[],
         merge=False,

diff --git a/db/scripts/upload/upload_experiment.py b/db/scripts/upload/upload_experiment.py
@@ -209,10 +209,10 @@ def create_correlation(
         create_relationship(
             source_file="tf_tg_corr.csv",
             type_="CORRELATION",
-            between=(("SYMBOL", "TF"), ("ENSEMBL", "ENSEMBL")),
+            between=(("ENSEMBL", "ENSEMBL_TF"), ("ENSEMBL", "ENSEMBL_TG")),
             node_types=("TF", "TG"),
-            values=["Correlation", "Source"],
-            reformat_values=[("Correlation", "toFloat"), ("Source", "toInteger")],
+            values=["Correlation", "p", "Source"],
+            reformat_values=[("Correlation", "toFloat"), ("Source", "toInteger"), ("p", "toFloat")],
             merge=False,
             driver=driver,
         )
@@ -225,8 +225,8 @@ def create_correlation(
             type_="CORRELATION",
             between=(("id", "id"), ("ENSEMBL", "ENSEMBL")),
             node_types=("OR", "TG"),
-            values=["Correlation", "Source"],
-            reformat_values=[("Correlation", "toFloat"), ("Source", "toInteger")],
+            values=["Correlation", "p", "Source"],
+            reformat_values=[("Correlation", "toFloat"), ("Source", "toInteger"), ("p", "toFloat")],
             merge=False,
             driver=driver,
         )