src/db_utils/creation_of_exon_table.py: major modification, this script will...

src/db_utils/creation_of_exon_table.py: major modification, this script will create and fill the cin_exon and cin_gene table of the new database model

src/db_utils/creation_of_exon_table.py

@@ -8,67 +8,83 @@ Description: Create the exon_ctrl file and the exon file
 
 from pathlib import Path
 import pandas as pd
-from typing import List
-from .config import Config, logging_def
+from ..logging_conf import logging_def
+from .config import Config
 import logging
+from .populate_database import populate_df
 
 
-def get_ctrl_exon(mfile: Path) -> List[str]:
+def get_genomic_region(bed_file: Path) -> pd.DataFrame:
     """
-    Get the control exons in mfile.
+    Create a file containing the control exons
 
-    :param mfile: A file containing control exons
-    :return: The lis of control exons in mfile
+    :param bed_file: A bed file containing intern exons
+    :return: the bed file as a dataframe
     """
-    with mfile.open("r") as f:
-        exon_list = f.read().splitlines()
-    return exon_list
+    df = pd.read_csv(bed_file, sep="\t")
+    df.columns = ["chromosome", "start", "stop", "id", "score", "strand"]
+    df = df.loc[(df.stop - df.start > 2), :]
+    return df
 
 
-def get_ctrl(bed_file: Path) -> None:
+def get_gene_table() -> pd.DataFrame:
     """
-    Create a file containing the control exons
+    Create the gene table.
 
-    :param bed_file: A bed file containing intern exons
-    :return: the bed file as a dataframe
+    :return: The gene_table.
     """
+    logging.debug("Load the bed file of genes")
+    df_gene = get_genomic_region(Config.bed_gene)
+    df_gene.columns = ["chromosome", "start", "stop", "id", "name", "strand"]
+    logging.debug(df_gene.head())
+    return df_gene
 
-    df = pd.read_csv(bed_file, sep="\t")
-    df.columns = ["chr", "start", "stop", "name", "score", "strand"]
-    df = df.loc[(df.stop - df.start > 2), :]
-    df = df.loc[:, "name"]
-    df.to_csv(Config.ctrl_exon_file, header=False, index=False)
+
+def get_exon_table(df_gene_name: pd.DataFrame) -> pd.DataFrame:
+    """
+    Create the exon_table
+
+    :param df_gene_name: A datframe of gene id and gene name
+    :return: The exon table
+    """
+    df_gene_name.columns = ['id_gene', 'gene_name']
+    df_gene_name['id_gene'].astype(int)
+    logging.debug("Load the bed file of exons")
+    df_exon = get_genomic_region(Config.bed_exon)
+    df_exon = df_exon.loc[df_exon['stop'] - df_exon['start'] > 2, :]
+    logging.debug(df_exon.head())
+    df_exon.drop('score', inplace=True, axis=1)
+    logging.debug('Creating id_gene column ...')
+    df_exon['id_gene'] = df_exon['id'].str.replace(r'_\d+$', '')
+    df_exon['id_gene'] = df_exon['id_gene'].astype(int)
+    logging.debug(df_exon.head())
+    logging.debug("Creation of a column pos")
+    df_exon['pos'] = df_exon['id'].str.replace(r'^\d+_', '')
+    df_exon['pos'] = df_exon['pos'].astype(int)
+    logging.debug(df_exon.head())
+    logging.debug('Merging with gene name dataframe')
+    df_exon = df_exon.merge(df_gene_name, how="left", on='id_gene')
+    df_exon['name'] = df_exon['gene_name'] + '_' + \
+        df_exon['pos'].astype('str')
+    df_exon.drop('gene_name', inplace=True, axis=1)
+    return df_exon
 
 
-def get_exon_table(ctrl_exon: Path, gene_tab: Path,
-                   logging_level: str = "DISABLE"):
+def main_fill_exon_n_gene(logging_level: str = "DISABLE") -> None:
     """
-    Create the table of exon.
+    Create the cin_exon and cin_gene table and insert them into \
+    the ChIA-PET dabatbase.
 
-    :param ctrl_exon: A file containing control exons
-    :param gene_tab:  A file containing FasterDB gene display \
-    in two columns: gene_symbol and fasterdb id.
-    :param logging_level: The level of display
+    :param logging_level: The level of information to display
     """
-    logging_def(Config.output, logging_level)
-    ctrl_exon_list = get_ctrl_exon(ctrl_exon)
-    logging.debug("Creating dataframe of control exons")
-    df_exon = pd.DataFrame({"id": ctrl_exon_list})
-    logging.debug(f"{df_exon.head()}\nshape:{df_exon.shape}")
-    logging.debug("Creation of a pos column")
-    df_exon["pos"] = df_exon["id"].apply(lambda x: int(x.split("_")[1]))
-    logging.debug(f"{df_exon.head()}\nshape:{df_exon.shape}")
-    logging.debug("Creation of a gene_id column")
-    df_exon["gene_id"] = df_exon["id"].apply(lambda x: int(x.split("_")[0]))
-    logging.debug(f"{df_exon.head()}\nshape:{df_exon.shape}")
-    logging.debug("Addition of a gene column")
-    gene_df = pd.read_csv(gene_tab, sep="\t")
-    gene_df.columns = ["gene", "gene_id"]
-    df_exon = df_exon.merge(gene_df)
-    logging.debug(f"{df_exon.head()}\nshape:{df_exon.shape}")
-    logging.debug("Addition of a name column")
-    df_exon["name"] = df_exon.apply(lambda x: f"{x.gene}_"
-                                              f"{x.pos}", axis=1)
-    df_exon = df_exon[["pos", "name", "id", "gene_id"]]
-    logging.debug(f"{df_exon.head()}\nshape:{df_exon.shape}")
-    df_exon.to_csv(Config.exon_file, sep="\t", index=False)
+    logging_def(Config.db_file.parent, __file__, logging_level)
+    df_gene = get_gene_table()
+    df_exon = get_exon_table(df_gene[['id', 'name']])
+    logging.debug('Filling cin_gene table')
+    populate_df(table='cin_gene', df=df_gene, clean='y')
+    logging.debug('Filling cin_exon table')
+    populate_df(table='cin_exon', df=df_exon, clean='y')
+
+
+if __name__ == '__main__':
+    main_fill_exon_n_gene('DEBUG')