This tool is designed to predict the oncogenic potential of fusion genes found by Next-Generation Sequencing in cancer cells. It is a post-processing step that tries to validate in-silico the predictions made by fusion detection software. Oncofuse is NOT a fusion detection software, its goal is NOT to identify fusion sequences, but to assign a functional prediction score (oncogenic potential, i.e. the probability of being 'driver' events) to fusion sequences identified by other software such as Tophat-fusion, fusioncatcher or STAR.

官网地址：https://genetica.unav.edu/oncofuse.html

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
# @Date    : 2023-7-13
# @Author  : lizhengnan

import os
import argparse
import sys

parser = argparse.ArgumentParser(description="This script performs Oncofuse analysis")
parser.add_argument("fusion_result",
                    help="Path to the fusion result file, example: /TJPROJ6/RNA_SH/shouhou/202307/X101SC23022823-Z01-J004/03.Result_X101SC23022823-Z01-J004-B4-16_Homo_sapiens/Result_X101SC23022823-Z01-J004-B4-16/8.Fusion/1.fusionlist/")
parser.add_argument("sample2tissue",
                    help="Path to the sample to tissue file, the four supported tissue types: EPI (epithelial origin), HEM (hematological origin), MES (mesenchymal origin) and AVG (average expression, if tissue source is unknown)")
#parser.add_argument("sample", help="sample name, split by ','")
parser.add_argument("--result_dir", help="Path to the directory to save the results", nargs='?', default=None)
parser.add_argument("--genome_version", help="Genome assembly version, default is hg38. Allowed values: hg18, hg19, hg38", default='hg38')
args = parser.parse_args()

fusion_result = args.fusion_result.strip()
sample2tissue = args.sample2tissue.strip()
#samples = args.sample.strip().split(',')

if args.result_dir.strip() is None :
    result_dir = args.result_dir.strip()
else:
    result_dir = os.path.join(os.getcwd(), 'result')
if not os.path.exists(result_dir):
    os.mkdir(result_dir)

if args.genome_version.strip() :
    genome_version = args.genome_version.strip()
    
# TISSUE
TISSUE = ['EPI', 'HEM', 'MES', 'AVG']
sample2tissue_dict = {}
with open(sample2tissue) as f:
    sample2tissue_tmp = f.readlines()
for i in sample2tissue_tmp:
    if i.startswith('sample'):
        continue
    else:
        sample = i.strip().split('\t')[0]
        tissue = i.strip().split('\t')[1]
        if tissue not in TISSUE:
            tissue = 'AVG'
        else:
            pass
        sample2tissue_dict[sample] = tissue

# coord
samples = sample2tissue_dict.keys()
coord_dir = os.path.join(os.getcwd(), 'coord')
if not os.path.exists(coord_dir):
    os.mkdir(os.path.join(os.getcwd(), 'coord'))

for i in samples:
    tmp_fusion_result = open(fusion_result+"/"+i+"_fusion.xls").readlines()
    header = tmp_fusion_result.pop(0).strip().split('\t')
    LeftBreakpoint_index = header.index('LeftBreakpoint')
    RightBreakpoint_index = header.index('RightBreakpoint')
    JunctionReadCount_index = header.index('JunctionReadCount')
    SpanningFragCount_index = header.index('SpanningFragCount')

    with open(coord_dir + "/" + i, 'w') as tmp_out:
        tmp_out.write("\t".join(["#5' chrom", "5' coord", "3' chrom", "3' coord", "tissue", "5' FPG", "3' FPG", "spanning", "encompassing"])+"\n")

    for l in tmp_fusion_result:
        line = l.strip().split('\t')
        with open(coord_dir+"/"+i, 'a') as tmp_out:
            tmp_out.write("\t".join(line[LeftBreakpoint_index].split(":")[0:2]) + "\t" +
                          "\t".join(line[RightBreakpoint_index].split(":")[0:2]) + "\t" + sample2tissue_dict[i] + "\t" +
                          line[LeftBreakpoint_index].split(":")[2] + "\t" + line[RightBreakpoint_index].split(":")[2] +
                          "\t" + line[JunctionReadCount_index] + "\t" + str(int(line[JunctionReadCount_index])+int(line[SpanningFragCount_index]))+"\n")

# Script

cmd = '''#!/bin/bash\n
export java="/PUBLIC/software/public/System/jre1.8.0_25/bin/java"\n\n
cd {}\n
'''.format(coord_dir)

for i in samples:
    cmd += 'java -jar /TJPROJ6/RNA_SH/software/Oncofuse/oncofuse-1.1.1/Oncofuse.jar -a {} -p 2  {} coord - {}/{}_Oncofuse_result.xls\n\n'.format(genome_version, i, result_dir, i)

cmd += 'cp /TJPROJ6/RNA_SH/software/Oncofuse/oncofuse-1.1.1/readme.txt {}'.format(result_dir)

with open(os.getcwd() + "/oncofuse.sh", 'w') as script:
    script.write(cmd)

os.system("qsub -V -cwd -l vf=4G,p=2 {}".format(os.getcwd() + "/oncofuse.sh"))

python Oncofuse.py  --help
usage: Oncofuse.py [-h] [--result_dir [RESULT_DIR]]
                   [--genome_version GENOME_VERSION]
                   fusion_result sample2tissue

This script performs Oncofuse analysis

positional arguments:
  fusion_result         Path to the fusion result file, example: /TJPROJ6/RNA_
                        SH/shouhou/202307/X101SC23022823-Z01-J004/03.Result_X1
                        01SC23022823-Z01-J004-B4-16_Homo_sapiens/Result_X101SC
                        23022823-Z01-J004-B4-16/8.Fusion/1.fusionlist/
  sample2tissue         Path to the sample to tissue file, the four supported
                        tissue types: EPI (epithelial origin), HEM
                        (hematological origin), MES (mesenchymal origin) and
                        AVG (average expression, if tissue source is unknown)

optional arguments:
  -h, --help            show this help message and exit
  --result_dir [RESULT_DIR]
                        Path to the directory to save the results
  --genome_version GENOME_VERSION
                        Genome assembly version, default is hg38. Allowed
                        values: hg18, hg19, hg38

/TJPROJ6/RNA_SH/software/Oncofuse/oncofuse-1.1.1/test