def test_1(self):
agfusion_db = agfusion.AGFusionDB("agfusion.homo_sapiens.84.db", debug=False)
all_fusions = ['Adamts9-Ano2','Trp53-Sat2','1700112E06Rik-Runx1','Runx1-1700112E06Rik','Rell1-Lhfpl3','Phc1-Smarca2','Lrrc8d-Gbp11','C920009B18Rik-H60b']
for fusion in agfusion.parsers['fusioncatcher']('./data/FusionsFindingAlgorithms/FusionCatcher/final-list_candidate-fusion-genes.txt',db.logger):
fusion = agfusion.Fusion(
gene5prime=fusion['gene5prime'],
gene5primejunction=fusion['gene5prime_junction'],
gene3prime=fusion['gene3prime'],
gene3primejunction=fusion['gene3prime_junction'],
db=db,
pyensembl_data=data,
protein_databases=['pfam'],
noncanonical=False
)
assert fusion.name in all_fusions, '%s not in list!' % fusion.name
def main():
"""
Main function for processing command line options
"""
parser = argparse.ArgumentParser(
description='Annotate Gene Fusion (AGFusion)')
subparsers = parser.add_subparsers(help='AGFusion programs.',
dest="subparser_name")
annotate_parser = subparsers.add_parser(
'annotate', help='Annotate and visualize a single fusion.')
annotate_parser.add_argument('-g5',
'--gene5prime',
type=str,
required=True,
help='5\' gene partner')
annotate_parser.add_argument('-g3',
'--gene3prime',
type=str,
required=True,
help='3\' gene partner')
annotate_parser.add_argument(
'-j5',
'--junction5prime',
type=int,
required=True,
help='Genomic location of predicted fuins for the 5\' gene partner. ' +
'The 1-based position that is the last nucleotide included in ' +
'the fusion before the junction.')
annotate_parser.add_argument(
'-j3',
'--junction3prime',
type=int,
required=True,
help='Genomic location of predicted fuins for the 3\' gene partner. ' +
'The 1-based position that is the first nucleotide included in ' +
'the fusion after the junction.')
add_common_flags(annotate_parser)
annotate_parser.add_argument(
'--scale',
type=int,
required=False,
default=-1,
help='(Optional) Set maximum width (in amino acids) of the ' +
'figure to rescale the fusion (default: max length of ' +
'fusion product)')
# batch file parser
batch_parser = subparsers.add_parser(
'batch',
help='Annotate fusions from an output file from a fusion ' +
'finding algorithm.')
batch_parser.add_argument(
'-f',
'--file',
type=str,
required=True,
help='Output file from fusion-finding algorithm.')
batch_parser.add_argument(
'-a',
'--algorithm',
type=str,
required=True,
help='The fusion-finding algorithm. Can be one of the following: ' +
', '.join(agfusion.parsers.keys()) + '.')
add_common_flags(batch_parser)
# download database
database_parser = subparsers.add_parser(
'download', help='Download database for a reference genome.')
database_parser.add_argument(
'-d',
'--dir',
type=str,
default='',
help='(Optional) Directory to the database will be downloaded ' +
'to (defaults to current working directory).')
database_parser.add_argument(
'-g',
'--genome',
type=str,
default=None,
help='Specify the genome shortcut (e.g. hg19). To see all' +
'available shortcuts run \'agfusion download -a\'. Either ' +
'specify this or --species and --release.')
database_parser.add_argument('-s',
'--species',
type=str,
default=None,
help='The species (e.g. homo_sapiens).')
database_parser.add_argument('-r',
'--release',
type=int,
default=None,
help='The ensembl release (e.g. 87).')
database_parser.add_argument(
'-a',
'--available',
action='store_true',
required=False,
help='List available species and ensembl releases.')
# build database parser
build_database_parser = subparsers.add_parser(
'build', help='Build database for a reference genome.')
build_database_parser.add_argument(
'-d',
'--dir',
type=str,
required=True,
help='Directory to write database file to.')
build_database_parser.add_argument('-s',
'--species',
type=str,
required=True,
help='The species (e.g. homo_sapiens).')
build_database_parser.add_argument('-r',
'--release',
type=int,
required=True,
help='The ensembl release (e.g. 87).')
build_database_parser.add_argument(
'--pfam',
type=str,
required=True,
help='File containing PFAM ID mappings.')
build_database_parser.add_argument(
'--server',
type=str,
required=False,
default='ensembldb.ensembl.org',
help='(optional) Ensembl server (default ensembldb.ensembl.org)')
# agfusion version number
parser.add_argument('-v',
'--version',
action='version',
version=agfusion.__version__)
args = parser.parse_args()
if args.subparser_name == 'build':
builddb(args)
exit()
elif args.subparser_name == 'download':
if args.available:
list_available_databases()
else:
downloaddb(args)
exit()
# single or batch mode
if not exists(args.out):
mkdir(args.out)
# if user does not specify a sqlite database then use the one provided
# by the package
db_file = split(args.database)[1]
species = db_file.split('.')[1]
release = db_file.split('.')[2]
assert species in AVAILABLE_ENSEMBL_SPECIES, 'unsupported species!'
agfusion_db = agfusion.AGFusionDB(args.database, debug=args.debug)
agfusion_db.build = species + '_' + str(release)
# get the pyensembl data
pyensembl_data = pyensembl.EnsemblRelease(release, species)
try:
pyensembl_data.db
except ValueError:
agfusion_db.logger.error(
"Missing pyensembl data. Run pyensembl install --release " +
"{} --species {}".format(release, species))
exit()
# parse the re-coloring and re-naming
colors = {}
rename = {}
if args.rename is not None:
for i in args.rename:
pair = i.split(';')
assert len(pair) == 2, " did not properly specify --rename"
if pair[0] in rename:
agfusion_db.logger.warn(
"WARNING - you rename {} twice.".format(pair[0]))
rename[pair[0]] = pair[1]
if args.recolor is not None:
for i in args.recolor:
pair = i.split(';')
assert len(pair) == 2, " did not properly specify --colors"
if pair[0] in colors:
agfusion_db.logger.warn(
"You specified colors for {} twice.".format(pair[0]))
if pair[0] in rename:
colors[rename[pair[0]]] = pair[1]
else:
colors[pair[0]] = pair[1]
# check image file type is valid
if args.type not in ['png', 'pdf', 'jpeg']:
agfusion_db.logger.error(
"ERROR - provided an incorrect image file type: {}.".format(
args.type))
exit()
if args.subparser_name == 'annotate':
annotate(gene5prime=args.gene5prime,
junction5prime=args.junction5prime,
gene3prime=args.gene3prime,
junction3prime=args.junction3prime,
agfusion_db=agfusion_db,
pyensembl_data=pyensembl_data,
args=args,
outdir=args.out,
colors=colors,
rename=rename,
scale=args.scale)
elif args.subparser_name == 'batch':
batch_mode(args, agfusion_db, pyensembl_data, rename, colors)
from os.path import join, expanduser, curdir, abspath
import unittest
import agfusion
from agfusion import utils
import pyensembl
from Bio import SeqIO
data = pyensembl.EnsemblRelease(84, 'mouse')
db = agfusion.AGFusionDB(abspath(join(curdir, 'agfusion.mus_musculus.84.db')))
db.build = 'mus_musculus_84'
class TestSequencePrediction(unittest.TestCase):
def test_1(self):
"""
test CDS and cDNA correct for junction that is on exon boundaries and
produces an in-frame protein.
"""
#test the dna and protein coding sequences are correct by comparing
#with manually generally sequences
fusion = agfusion.Fusion(gene5prime="ENSMUSG00000022770",
gene5primejunction=31684294,
gene3prime="ENSMUSG00000002413",
gene3primejunction=39648486,
db=db,
pyensembl_data=data,
protein_databases=['pfam', 'tmhmm'],
noncanonical=True)
from os.path import join, expanduser, curdir, abspath
import unittest
import agfusion
from agfusion import utils
import pyensembl
from Bio import SeqIO, Seq, Alphabet
data = pyensembl.EnsemblRelease(84,'mouse')
db = agfusion.AGFusionDB(abspath(join(curdir,'agfusion.mus_musculus.84.db')))
db.build = 'mus_musculus_84'
data_human = pyensembl.EnsemblRelease(75,'human')
db_human = agfusion.AGFusionDB(abspath(join(curdir,'agfusion.homo_sapiens.75.db')))
db_human.build = 'homo_sapiens_75'
class TestSequencePrediction_human(unittest.TestCase):
def test_1(self):
"""
test CDS and prortein correct for junction that is on exon boundaries and
produces an out-of-frame protein.
"""
#test the dna and protein coding sequences are correct by comparing
#with manually generally sequences
fusion = agfusion.Fusion(
gene5prime="TMEM87B",
gene5primejunction=112843681,
gene3prime="MERTK",
gene3primejunction=112722768,
from flask_errormail import mail_on_500
parser = argparse.ArgumentParser(description='AGFusion web application')
parser.add_argument('--database',
type=str,
required=True,
help='Path to the database file (e.g. agfusion.db)')
args = parser.parse_args()
ADMINISTRATORS = ('*****@*****.**')
app = Flask(__name__)
app.config.from_object(__name__)
mail_on_500(app, ADMINISTRATORS)
db = agfusion.AGFusionDB(args.database)
app.config.update(
dict(DATABASE=None,
SECRET_KEY='development key',
USERNAME='******',
PASSWORD='******'))
app.config.from_envvar('FLASKR_SETTINGS', silent=True)
def check_params(params):
#check parameters
if params['dpi'] == '':
params['dpi'] = '100'