diff --git a/src/db_utils/db_creation.py b/src/db_utils/db_creation.py
index ccfed76ee3fd0616cf19ad91ab26ba31d9caa572..079008273fc48ebd3090af963a9a4bbf37237e7a 100755
--- a/src/db_utils/db_creation.py
+++ b/src/db_utils/db_creation.py
@@ -124,6 +124,7 @@ def create_cin_exon_interaction_table(conn: sqlite3.Connection) -> None:
[exon2] VARCHAR(30) NOT NULL,
[id_project] INT NOT NULL,
[level] VARCHAR(25) NOT NULL,
+ [distance] VARCHAR(30),
PRIMARY KEY ([id]),
FOREIGN KEY ([exon1]) REFERENCES cin_exon([id]),
FOREIGN KEY ([exon2]) REFERENCES cin_exon([id]),
@@ -146,6 +147,7 @@ def create_cin_gene_interaction_table(conn: sqlite3.Connection) -> None:
[gene2] INT NOT NULL,
[id_project] INT NOT NULL,
[level] VARCHAR(25) NOT NULL,
+ [distance] VARCHAR(30),
PRIMARY KEY ([id]),
FOREIGN KEY ([gene1]) REFERENCES cin_gene([id]),
FOREIGN KEY ([gene2]) REFERENCES cin_gene([id]),
diff --git a/src/db_utils/interactions/config_interactions.py b/src/db_utils/interactions/config_interactions.py
index 753a9a2c3b2d238b5bd6a45ae8388d030b03812b..517923d9da0d4703462d2e3e981de3feb952e539 100644
--- a/src/db_utils/interactions/config_interactions.py
+++ b/src/db_utils/interactions/config_interactions.py
@@ -8,6 +8,7 @@ Description: Configuration variables for subfolder interactions.
"""
from ..config import Config
+from pathlib import Path
class ConfigInteractions:
@@ -30,3 +31,5 @@ class ConfigInteractions:
pet_vs_gene_output = chia_pet_interaction / 'intersections_gene'
couple_exon = chia_pet_interaction / 'couple_exon'
couple_gene = chia_pet_interaction / 'couple_gene'
+ results = Path(__file__).parents[3] / "results"
+ db_file = results / 'chia_pet_database.db'
diff --git a/src/db_utils/interactions/features_interactions.py b/src/db_utils/interactions/features_interactions.py
index 6369dbb85172f6815b964c1184abad4d5ea01a5f..92253d8be30ee66ed26948b8eb2828daa4c5d35f 100644
--- a/src/db_utils/interactions/features_interactions.py
+++ b/src/db_utils/interactions/features_interactions.py
@@ -30,6 +30,7 @@ import re
import numpy as np
import os.path
import sys
+import sqlite3
from ..populate_database import populate_df
@@ -251,12 +252,91 @@ def filtering_2(df_filter_2: pd.DataFrame):
if FILE_1 == FILE_2:
df_filter_2["id_project"] = FILE_1
df_filter_2["id2"] = df_filter_2["id"].astype(str) + "_" + \
- df_filter_2["id_project"].astype(str)
+ df_filter_2["id_project"].astype(str)
del df_filter_2["id"]
df_filter_2.rename(columns={"id2": "id"}, inplace=True)
return df_filter_2
+def get_info_from_database(cnx: sqlite3.Connection, query: str) -> \
+ pd.DataFrame:
+ """
+ Get the exons and the genes information from the database, for example:
+ - for the exons: 1_1 18 28681865 28682388
+ - for the genes: 1 18 28645943 28682388
+
+ :param cnx: connexion to the ChIA-PET database
+ :param query: the SQL query that allows us to get data from the database
+ :return df_res: the dataframe with the data obtained
+ """
+ cursor = cnx.cursor()
+ cursor.execute(query)
+ res = list(cursor.fetchall())
+ df_res = pd.DataFrame(res, columns=["ID", "chr", "start", "stop"])
+ return df_res
+
+
+def add_info_distance_between_features(df: pd.DataFrame) -> pd.DataFrame:
+ """
+ Allows to calculate the distance between the two genes or the two exons
+ studied in interaction and to add this information in the result dataframe.
+ If the result is NULL it is because we study two exons or genes located in
+ the same chromosome.
+ If the result is (null), so NaN, it is because we study two exons or genes
+ which have different identifiers, but strictly identical coordinates.
+
+ :param df: Result of the "filtering_2" function
+ :return df: df with distances added or NULL or (null) see before for more
+ details.
+ """
+ info_exon = get_info_from_database(sqlite3.connect(Config.db_file),
+ """SELECT id, chromosome, start, stop
+ FROM cin_exon""")
+ info_gene = get_info_from_database(sqlite3.connect(Config.db_file),
+ """SELECT id, chromosome, start, stop
+ FROM cin_gene""")
+ if OPT_EXON == "on":
+ df = df.merge(info_exon, left_on="exon1", right_on="ID")
+ df = df.rename(columns={'ID': 'ID1', 'chr': 'chr1', 'start': 'start1',
+ 'stop': 'stop1'})
+ df = df.merge(info_exon, left_on="exon2", right_on="ID")
+ df = df.rename(columns={'ID': 'ID2', 'chr': 'chr2', 'start': 'start2',
+ 'stop': 'stop2'})
+ df.loc[df["start1"] < df["start2"], "distance"] = \
+ (df["start2"] - df["stop1"] + 1)
+ df.loc[df["stop1"] < df["stop2"], "distance"] = \
+ (df["start2"] - df["stop1"] + 1)
+ df.loc[df["start1"] > df["start2"], "distance"] = \
+ (df["start1"] - df["stop2"] + 1)
+ df.loc[df["stop1"] > df["stop2"], "distance"] = \
+ (df["start1"] - df["stop2"] + 1)
+ df.loc[df["chr1"] != df["chr2"], "distance"] = "NULL"
+ df.drop(["start1", "ID1", "chr1", "stop1", "ID2", "chr2", "start2",
+ "stop2"], axis='columns', inplace=True)
+ elif OPT_GENE == "on":
+ info_gene["ID"] = info_gene["ID"].astype(str)
+ df = df.merge(info_gene, left_on="gene1", right_on="ID")
+ df = df.rename(columns={'ID': 'ID1', 'chr': 'chr1', 'start': 'start1',
+ 'stop': 'stop1'})
+ df = df.merge(info_gene, left_on="gene2", right_on="ID")
+ df = df.rename(columns={'ID': 'ID2', 'chr': 'chr2', 'start': 'start2',
+ 'stop': 'stop2'})
+ df.loc[df["start1"] < df["start2"], "distance"] = \
+ (df["start2"] - df["stop1"] + 1)
+ df.loc[df["stop1"] < df["stop2"], "distance"] = \
+ (df["start2"] - df["stop1"] + 1)
+ df.loc[df["start1"] > df["start2"], "distance"] = \
+ (df["start1"] - df["stop2"] + 1)
+ df.loc[df["start1"] > df["start2"], "distance"] = \
+ (df["start1"] - df["stop2"] + 1)
+ df.loc[df["stop1"] > df["stop2"], "distance"] = \
+ (df["start1"] - df["stop2"] + 1)
+ df.loc[df["chr1"] != df["chr2"], "distance"] = "NULL"
+ df.drop(["start1", "ID1", "chr1", "stop1", "ID2", "chr2",
+ "start2", "stop2"], axis='columns', inplace=True)
+ return df
+
+
def create_interaction_table(logging_level: str = "DISABLE"):
"""
Create the interaction tables.
@@ -281,6 +361,8 @@ def create_interaction_table(logging_level: str = "DISABLE"):
logging.debug("Sum weight of identical interaction")
df = filtering_2(df)
logging.debug(df.head())
+ df = add_info_distance_between_features(df)
+ logging.debug(df.head())
df.to_csv(MY_OUT_FILE, index=False, sep="\t", header=True)
if OPT_EXON == "on":
logging.debug("Filling cin_exon_interaction")