diff --git a/src/find_interaction_cluster/ppi_scripts/__init__.py b/src/find_interaction_cluster/ppi_scripts/__init__.py
new file mode 100644
index 0000000000000000000000000000000000000000..60f3a118d41cf7494b1cd067bf5d649b42ba1e05
--- /dev/null
+++ b/src/find_interaction_cluster/ppi_scripts/__init__.py
@@ -0,0 +1,7 @@
+#!/usr/bin/env python3
+
+# -*- coding: UTF-8 -*-
+
+"""
+Description:
+"""
diff --git a/src/find_interaction_cluster/ppi_scripts/__main__.py b/src/find_interaction_cluster/ppi_scripts/__main__.py
new file mode 100644
index 0000000000000000000000000000000000000000..f2f10e03ded5cbaf1644ba3a5f4883f1c0189483
--- /dev/null
+++ b/src/find_interaction_cluster/ppi_scripts/__main__.py
@@ -0,0 +1,20 @@
+#!/usr/bin/env python3
+
+# -*- coding: UTF-8 -*-
+
+"""
+Description:
+"""
+
+from .get_ensembl_hg19_gene_ccoordinates import create_protein_gene_file
+from .create_gene_interaction_file import get_protein_interaction_table
+
+def launcher_ppi():
+ """
+ launch every script in this submodules.
+ """
+ create_protein_gene_file()
+ get_protein_interaction_table()
+
+
+launcher_ppi()
\ No newline at end of file
diff --git a/src/find_interaction_cluster/ppi_scripts/config_ppi.py b/src/find_interaction_cluster/ppi_scripts/config_ppi.py
new file mode 100644
index 0000000000000000000000000000000000000000..312baea9d862909fa409ab7aed53984c2e50e702
--- /dev/null
+++ b/src/find_interaction_cluster/ppi_scripts/config_ppi.py
@@ -0,0 +1,25 @@
+#!/usr/bin/env python3
+
+# -*- coding: UTF-8 -*-
+
+"""
+Description: This script contains all the variables needed in this \
+submodules
+"""
+
+from ..config import ConfigGraph
+
+
+class ConfigPPI:
+ """
+ contains all the variables used in this submodules
+ """
+ output_folder = ConfigGraph.output_folder / "ppi_analysis"
+ data = ConfigGraph.data
+ bed_ensembl = data / "ppi_data" / "liftover_hg38gene2hg19.bed"
+ ensembl_gene = data / "ppi_data" / "mart_export_hg38_proteins_genes.txt"
+ ppi_string = data / "ppi_data" / "9606.protein.links.v11.0.txt.gz"
+ protein_gene = output_folder / "protein_gene.txt"
+ gene_ppi_file = output_folder / "ppi_gene_file.txt"
+ protein_gene_fasterdb = output_folder / "protein_gene_fasterdb.txt"
+ fasterdb_ppi = output_folder / "fasterdb_ppi.txt"
\ No newline at end of file
diff --git a/src/find_interaction_cluster/ppi_scripts/create_gene_interaction_file.py b/src/find_interaction_cluster/ppi_scripts/create_gene_interaction_file.py
new file mode 100644
index 0000000000000000000000000000000000000000..59221d693e4fa8d558d4b2018b94f8cc94964bfa
--- /dev/null
+++ b/src/find_interaction_cluster/ppi_scripts/create_gene_interaction_file.py
@@ -0,0 +1,175 @@
+#!/usr/bin/env python3
+
+# -*- coding: UTF-8 -*-
+
+"""
+Description: Create a file of genes having their proteins interacting with \
+each other
+"""
+
+from pathlib import Path
+from ..config import Config
+from .config_ppi import ConfigPPI
+import pandas as pd
+from typing import Dict, Optional, Tuple
+
+
+def load_string_ppi_file(ppi_file: Path) -> pd.DataFrame:
+ """
+ Load the string protein protein interaction file.
+
+ :param ppi_file: A file containing the protein protein interaction.
+ :return: The loaded string ppi interaction file
+ """
+ df = pd.read_csv(ppi_file, sep=" ")
+ df['protein1'] = df['protein1'].str.replace("9606.", "", regex=False)
+ df['protein2'] = df['protein2'].str.replace("9606.", "", regex=False)
+ return df
+
+
+def load_protein_gene_links(prot_gene_file: Path) -> Tuple[pd.DataFrame, Dict]:
+ """
+ Load the file linking each gene to it's hg19 coordinates and encoded \
+ proteins.
+
+ :param prot_gene_file: A file linking each gene to it's hg19 \
+ coordinates and encoded proteins.
+ :return: The loaded file and a dictionary of synonyms
+ """
+ df = pd.read_csv(prot_gene_file, sep="\t",
+ dtype={"Gene_stable_ID": str,
+ "Protein_stable_ID": str,
+ "Gene_name": str,
+ "Gene_Synonym": str,
+ "chr": str,
+ "start": int,
+ "stop": int,
+ "strand": str})
+ dic_synonym = {}
+ for i in range(df.shape[0]):
+ mseries = df.iloc[i, :]
+ if isinstance(mseries["Gene_Synonym"], float):
+ dic_synonym[mseries["Gene_name"].upper()] = \
+ mseries["Gene_name"].upper()
+ else:
+ dic_synonym[mseries["Gene_Synonym"].upper()] = \
+ mseries["Gene_name"].upper()
+ return df, dic_synonym
+
+
+def load_fasterdb_gene(bed_gene: Path, dic_synonym: Dict) -> Dict:
+ """
+ Load a bed file containing all fasterDB gene.
+
+ :param bed_gene: A bed file containing all fasterdb gene
+ :param dic_synonym: A dictionary of synonyms used to replace \
+ the old fasterDB names by the new ones
+ :return: A dictionary linking each fasterDB gene name to its coordinates
+ """
+ dic = {}
+ with bed_gene.open("r") as inbed:
+ for line in inbed:
+ line = line.replace("\n", "").split("\t")
+ gene_name = line[4].upper()
+ if gene_name in dic_synonym.keys():
+ gene_name = dic_synonym[gene_name]
+ if gene_name not in dic.keys():
+ dic[gene_name] = [[line[0], int(line[1]), int(line[2]),
+ int(line[3]), gene_name, line[5]]]
+ else:
+ dic[gene_name].append(
+ [line[0], int(line[1]), int(line[2]),
+ int(line[3]), gene_name, line[5]])
+ return dic
+
+
+def find_fasterdb_id(mseries: pd.Series, dic: Dict) -> Optional[int]:
+ """
+ Find the fasterDB gene id from a line corresponding to an ensembl gene.
+
+ :param mseries: A line containing the ensembl gene id and gene name of \
+ a gene
+ :param dic: A dictionary linking each gene name to it's coordinate
+ :return: The fasterDB id of the ensembl gene.
+ """
+ if mseries['Gene_name'].upper() not in dic.keys():
+ if isinstance(mseries['Gene_Synonym'], float) or \
+ mseries['Gene_Synonym'].upper() not in dic.keys():
+ return None
+ genes = dic[mseries['Gene_Synonym'].upper()]
+ else:
+ genes = dic[mseries['Gene_name'].upper()]
+ if len(genes) == 1:
+ return genes[0][3]
+ elif len(genes) >= 1:
+ for gene in genes:
+ if (
+ mseries['chr'] == gene[0]
+ and mseries['start'] == gene[1]
+ and mseries['stop'] == gene[2]
+ and mseries['strand'] == gene[5]
+ ):
+ return gene[3]
+ return None
+ return None
+
+
+def add_gene_id_column(prot_gene_df: pd.DataFrame, dic: Dict) -> pd.DataFrame:
+ """
+ Add the fasterDB gene id to ensembl gene then only return lines in \
+ the dataframe with a fasterDB gene id.
+
+ :param prot_gene_df: A table containing ensembl gene id.
+ :param dic: A dictionary linking gene name to is fasterdb id.
+ :return: The table prot_gene_df with a column FasterDB_id
+ """
+ prot_gene_df['FasterDB_id'] = prot_gene_df.apply(find_fasterdb_id,
+ axis=1, dic=dic)
+ df = prot_gene_df.loc[-prot_gene_df['FasterDB_id'].isna(), :].copy()
+ df['FasterDB_id'] = df['FasterDB_id'].astype(int)
+ df.drop(['Gene_Synonym'], axis=1, inplace=True)
+ return df.drop_duplicates()
+
+
+def add_fasterdb_id_to_ppi_file(ppi_df: pd.DataFrame,
+ df_fasterdb: pd.DataFrame) -> pd.DataFrame:
+ """
+ Add two columns to the ppi_df: one containing the fasterdb gene_id \
+ of the gene that encode the first interacting protein and the second \
+ containing the fasterdb gene id of the gene encoding the second \
+ interacting protein.
+
+ :param ppi_df: Dataframe of interaction
+ :param df_fasterdb: df with the fasterDB gene id of the gene en their \
+ encoding protein identified by ensembl protein id.
+ :return: The dataframe with the gene id of genes encoding the interacting \
+ protein.
+ """
+ df_fasterdb = df_fasterdb[["Protein_stable_ID", "FasterDB_id"]].copy()
+ df_fasterdb.rename({"FasterDB_id": "gene"}, axis=1, inplace=True)
+ ppi_df = ppi_df.merge(df_fasterdb, how="left", left_on="protein1",
+ right_on="Protein_stable_ID")
+ ppi_df = ppi_df.merge(df_fasterdb, how="left", left_on="protein2",
+ right_on="Protein_stable_ID", suffixes=['1', '2'])
+ ppi_df = ppi_df.loc[(-ppi_df['gene1'].isna()) &
+ (-ppi_df['gene2'].isna()), :]
+ ppi_df.drop(["Protein_stable_ID1", "Protein_stable_ID2"], axis=1,
+ inplace=True)
+ ppi_df['gene1'] = ppi_df['gene1'].astype(int)
+ ppi_df['gene2'] = ppi_df['gene2'].astype(int)
+ return ppi_df
+
+
+def get_protein_interaction_table():
+ """
+ link ensemble gene and their encoding protein to a fasterDB gene id.
+ """
+ df_prot_gene, dic_synonyms = \
+ load_protein_gene_links(ConfigPPI.protein_gene)
+ dic_gene = load_fasterdb_gene(Config.bed_gene, dic_synonyms)
+ df_fasterdb_id = add_gene_id_column(df_prot_gene, dic_gene)
+ df_fasterdb_id.to_csv(ConfigPPI.protein_gene_fasterdb, sep="\t",
+ index=False)
+ ppi_df = load_string_ppi_file(ConfigPPI.ppi_string)
+ ppi_fasterdb = add_fasterdb_id_to_ppi_file(ppi_df, df_fasterdb_id)
+ ppi_fasterdb.to_csv(ConfigPPI.fasterdb_ppi, sep="\t", index=False)
diff --git a/src/find_interaction_cluster/ppi_scripts/get_ensembl_hg19_gene_ccoordinates.py b/src/find_interaction_cluster/ppi_scripts/get_ensembl_hg19_gene_ccoordinates.py
new file mode 100644
index 0000000000000000000000000000000000000000..a2ecaeda470c7c56d08d951d1a9bb7ec3c9f0255
--- /dev/null
+++ b/src/find_interaction_cluster/ppi_scripts/get_ensembl_hg19_gene_ccoordinates.py
@@ -0,0 +1,83 @@
+#!/usr/bin/env python3
+
+# -*- coding: UTF-8 -*-
+
+"""
+Description: The goal of this script is to get only coding gene from \
+ensemble with their protein id and their coordinate in hg19.
+"""
+
+from .config_ppi import ConfigPPI
+import pandas as pd
+from pathlib import Path
+import numpy as np
+
+
+def load_genes(ensembl_file: Path) -> pd.DataFrame:
+ """
+ Load a table containing all the human protein of ensembl database \
+ and the name and location of their genes.
+
+ :param ensembl_file: A file containing all the human protein of \
+ ensembl database and the name and location of their genes.
+ :return: table containing all the human protein of ensembl database \
+ and the name and location of their genes.
+ """
+ df = pd.read_csv(ensembl_file, sep="\t",
+ dtype={"Gene_stable_ID": str,
+ "Gene_stable_ID_version": str,
+ "Protein_stable_ID": str,
+ "Chromosome-scaffold_name": str,
+ "Gene_start": int,
+ "Gene_end": int,
+ "Strand": str,
+ "Gene_name": str,
+ "Gene_Synonym": str})
+ df = df.loc[-df['Protein_stable_ID'].isna(), :]
+ return df[["Gene_stable_ID", "Protein_stable_ID", "Gene_name",
+ "Gene_Synonym"]]
+
+
+def load_bed(ensembl_bed: Path) -> pd.DataFrame:
+ """
+ Load a bed file containing ensembl hg38 gene on hg19 coordinates.
+
+ :param ensembl_bed: A bed file containing ensembl gene
+ :return: the bed file
+ """
+ df = pd.read_csv(ensembl_bed, names=["chr", "start", "stop", "name",
+ "score", "strand"], sep="\t")
+ df['chr'] = df["chr"].str.replace("chr", "")
+ df = df.drop_duplicates()
+ names = np.unique(df['name'].values, return_counts=True)
+ good_names = [cname for i, cname in enumerate(list(names[0]))
+ if names[1][i] == 1]
+ return df.loc[df['name'].isin(good_names), :]
+
+
+def merge_tables(df_gene: pd.DataFrame, df_bed: pd.DataFrame) -> pd.DataFrame:
+ """
+ Merges `df_gene` and `df_bed`together.
+
+ :param df_gene: table containing all the human protein of ensembl \
+ database and the name and location of their genes.
+ :param df_bed: A bed file containing ensembl gene
+ :return: The merged table
+ """
+ df_gene = df_gene.loc[
+ df_gene["Gene_name"].isin(list(df_bed['name'].unique())), :]
+ df_bed.rename({'name': "Gene_name"}, inplace=True, axis=1)
+ return df_gene.merge(df_bed, how="left", on="Gene_name")
+
+
+def create_protein_gene_file():
+ """
+ Create a file linking each ensembl protein to it's gene and the hg19 \
+ coordinate of this gene.
+ """
+
+ df_gene = load_genes(ConfigPPI.ensembl_gene)
+ df_bed = load_bed(ConfigPPI.bed_ensembl)
+ df = merge_tables(df_gene, df_bed)
+ ConfigPPI.protein_gene.parent.mkdir(parents=True, exist_ok=True)
+ df.to_csv(ConfigPPI.protein_gene, sep="\t", index=False)