def main(proj_path, snpEff_path):
proj_path = proj_path.rstrip('/')
snpEff_path = snpEff_path.rstrip('/')
write_log('Setting up SnpEff database for project reference genome',
proj_path)
name, version, genome, exons = load_db_tag(proj_path)
db_name = '{}.{}'.format(name, version)
create_dirs(snpEff_path, proj_path, db_name)
copy_rename(snpEff_path, proj_path, db_name, genome, exons)
update_conf(snpEff_path, proj_path, name, version)
build_db(snpEff_path, proj_path, db_name, exons)
write_log('New database set-up complete', proj_path)
def copy_rename(
path_to_snpEff, proj_path, db_name, genome, exons
): ### Copies and renames ref.fa and associated .gff or .gtf files to snpEff directory ###
write_log('Preparing to copy reference files', proj_path)
exons_ext = exons.split('.')[-1]
if 'gff' in exons_ext: exons_ext = 'gff'
else: exons_ext = 'gtf'
new_exons = 'genes.{}'.format(exons_ext)
check = '{}/data/{}/{}'.format(path_to_snpEff, db_name, new_exons)
try:
os.stat(check)
write_log('{} already exists'.format(check), proj_path)
except:
write_log('Copying {} to {}'.format(exons, check), proj_path)
shutil.copy('{}/reference/{}'.format(proj_path, exons), check)
new_genome = '{}.{}'.format(db_name, genome.split('.')[-1])
check = '{}/data/genomes/{}'.format(path_to_snpEff, new_genome)
try:
os.stat(check)
write_log('{} already exists'.format(check), proj_path)
except:
write_log('Copying {} to {}'.format(genome, check), proj_path)
shutil.copy('{}/reference/{}'.format(proj_path, genome), check)
def build_db(path_to_snpEff, proj_path, db_name,
genes): ### Runs snpEff build database ###
if 'snpEffPredictor.bin' in os.listdir('{}/data/{}'.format(
path_to_snpEff, db_name)):
write_log('Database already exists', proj_path)
return
write_log('Building SnpEff databade', proj_path)
if 'gff' in genes.split('.')[-1]: setting = 'gff3'
else: setting = 'gtf22'
command = 'java -jar {}/snpEff.jar build -{} -v {}'.format(
path_to_snpEff, setting, db_name)
print(command)
pipe = subprocess.PIPE
proc = subprocess.run(command, shell=True, stdout=pipe, stderr=pipe)
if proc.stdout:
write_log('[stdout]\n{}'.format(proc.stdout.decode()), proj_path)
if proc.stderr:
write_log('[stderr]\n{}'.format(proc.stderr.decode()), proj_path)
def create_dirs(
path_to_snpEff, proj_path, db_name
): ### Creates data/db_name and data/genomes directories if necessary ###
write_log('Preparing to create directorires', proj_path)
paths = ('{}/data/{}'.format(path_to_snpEff, db_name),
'{}/data/genomes'.format(path_to_snpEff))
for path in paths:
if not os.path.exists(path):
write_log('Creating directory {}'.format(path), proj_path)
os.makedirs(path)
else:
write_log('{} already exists.'.format(path), proj_path)
def update_conf(
path_to_snpEff, proj_path, name, version
): ### Checks for new database in snpEff.config, adds if not found ###
write_log('Preparing to update SnpEff.config', proj_path)
filepath = '{}/snpEff.config'.format(path_to_snpEff.rstrip('/'))
str_to_add = '\n# {name} genome, version {version}\n{name}.{version}.genome : N/A\n'.format(
name=name, version=version)
with open(filepath, 'r+') as f:
file = f.read()
if str_to_add in file: # re.search(str_to_add, file)
write_log('SnpEff.config already contains db tag', proj_path)
else:
write_log('Adding database to SnpEff.config', proj_path)
f.write(str_to_add)