diff --git a/src/find_interaction_cluster/tad_to_communities/tad_to_communities.py b/src/find_interaction_cluster/tad_to_communities/tad_to_communities.py
index 2735d158d04e5db0827ae8eb3a16c039b4dabc09..ef51ec27243ba9b290aaeed168cc5562a8b55acc 100644
--- a/src/find_interaction_cluster/tad_to_communities/tad_to_communities.py
+++ b/src/find_interaction_cluster/tad_to_communities/tad_to_communities.py
@@ -26,9 +26,11 @@ import subprocess
from .config import Config
import os
import pandas as pd
+from pathlib import Path
+from typing import Dict, List
-def launch_bedtools_intersect():
+def launch_bedtools_intersect() -> None:
"""
Launch a bedtools intersect command, between a BED file with TAD
annotations and a BED file with fasterDB annotation of genes, in order to
@@ -46,7 +48,27 @@ def launch_bedtools_intersect():
stderr=subprocess.STDOUT)
-def obtain_tad_to_genes(my_file: str):
+def targeted_bedtools_intersect(tad_file: Path, feature: str = "exon") -> Path:
+ """
+ Launch a bedtools intersect command, between a BED file with TAD
+ annotations and a BED file with fasterDB annotation of genes or genes or \
+ exons.
+
+ :param tad_file: A bed file containing TAD
+ :param feature: The kind of feature to analyse
+ :return: The bed file created by bedtools
+ """
+ Config.intersect_tad_gene.mkdir(exist_ok=True, parents=True)
+ target = Config.target_gene if feature == "gene" else Config.target_exon
+ basename_tad = tad_file.name.replace(".bed", "")
+ outfile = Config.intersect_tad_gene / f"{basename_tad}_{feature}.bed"
+ cmd = f"bedtools intersect -F 0.5 -a {tad_file} -b {target} " \
+ f"-wa -wb > {outfile}"
+ subprocess.check_call(cmd, shell=True, stderr=subprocess.STDOUT)
+ return outfile
+
+
+def obtain_tad_to_genes(my_file: str) -> Dict[str, List]:
"""
Allows to obtain a dictionary with this format:
tad_to_gene[TAD_ID] = [list_of_TAD_genes],
@@ -58,19 +80,18 @@ def obtain_tad_to_genes(my_file: str):
e.g.: tad_to_gene[TAD_Stein_MCF7_1] = [13028, 13038]
"""
tad_to_gene = {}
- my_file_t = str(Config.intersect_tad_gene) + "/" + my_file
- with open(my_file_t, "r") as my_file:
+ with open(my_file, "r") as my_file:
for line in my_file:
lclean = line.strip()
elmt = lclean.split("\t")
if elmt[3] not in tad_to_gene:
tad_to_gene[elmt[3]] = [elmt[-3]]
- elif elmt[3] in tad_to_gene:
+ else:
tad_to_gene[elmt[3]] += [elmt[-3]]
return tad_to_gene
-def obtain_final_df(tad_to_gene: dict, my_file: str):
+def obtain_final_df(tad_to_gene: dict, my_file: str, outfile: str) -> None:
"""
Allows to obtain the output file, with this format:
community nodes genes project, e.g.:
@@ -82,20 +103,21 @@ def obtain_final_df(tad_to_gene: dict, my_file: str):
:param tad_to_gene: a dictionary with this format:
tad_to_gene[TAD_ID] = [list_of_TAD_genes], see obtain_tad_to_genes()
:param my_file: the file obtain as result of launch_bedtools_intersect()
+ :param outfile: The output file
"""
- content = []
- for key, value in tad_to_gene.items():
- content.append([key, ", ".join(value), len(value)])
+ content = [
+ [key, ", ".join(value), len(value)]
+ for key, value in tad_to_gene.items()
+ ]
+
df = pd.DataFrame(content, columns=["community", "genes", "nodes"])
df["project"] = my_file
df = df[["community", "nodes", "genes", "project"]]
Config.tad_communities.mkdir(exist_ok=True, parents=True)
- my_file_out = str(Config.tad_communities) + "/communities_" + \
- my_file.replace("_gene.bed", ".txt")
- df.to_csv(my_file_out, sep='\t', index=False)
+ df.to_csv(outfile, sep='\t', index=False)
-def main_tad_to_communities():
+def main_tad_to_communities() -> None:
"""
Allows to launch all the functions of tad_to_communities module:
launch_bedtools_intersect() - obtain_tad_to_genes() and obtain_final_df()
@@ -106,6 +128,46 @@ def main_tad_to_communities():
results = os.listdir(Config.intersect_tad_gene)
for my_file in results:
print("Produce output file for:", my_file)
- tad_to_gene = obtain_tad_to_genes(my_file)
- obtain_final_df(tad_to_gene, my_file)
+ tad_to_gene = obtain_tad_to_genes(str(Config.intersect_tad_gene) +
+ "/" + my_file)
+ outfile = str(Config.tad_communities) + "/communities_" + \
+ my_file.replace("_gene.bed", ".txt")
+ obtain_final_df(tad_to_gene, my_file, outfile)
print("Results are in:", Config.tad_communities)
+
+
+def removing_chr(tad_file: Path) -> Path:
+ """
+ Remove the heading chr in the first column of the bed file \
+ if it is present
+ :param tad_file: A bed file containing tad
+ :return: The file containing the tad but without 'chr'
+ """
+
+ df = pd.read_csv(tad_file, header=None, sep="\t")
+ df[0] = df[0].str.replace("chr", "")
+ outfile = Config.intersect_tad_gene / tad_file.name
+ df.to_csv(outfile, sep="\t", header=False, index=False)
+ return outfile
+
+
+def targeted_tad_to_communities(tad_file: str, outfile: str, feature: str
+ ) -> None:
+ """
+ Create a community file from a bed file containing TADs.
+
+ :param tad_file: A bed file containing TAD
+ :param outfile: The resulting community file
+ :param feature: The kind of feature to analyse
+ """
+ outfile = Path(outfile)
+ if not outfile.is_file():
+ tad_file = removing_chr(Path(tad_file))
+ bed_out = targeted_bedtools_intersect(tad_file, feature)
+ dic_gene = obtain_tad_to_genes(str(bed_out))
+ obtain_final_df(dic_gene, bed_out.name.replace(".bed", ""),
+ str(outfile))
+ bed_out.unlink()
+ tad_file.unlink()
+ else:
+ print(f"{outfile} already exists !")
\ No newline at end of file