def __init__(self,
queryid,
result_fn=None,
iteration=None,
use_iteration_info=True,
strict_id_match=False):
'''
HHsearch parses the HHsearch result file.
Its major functionality is parsing the alignments and
scores and return the GenericPairwiseAlignment object
So that the alignments and scores can be easily manipulated.
'''
self.queryid = queryid
self.result_fn = result_fn
self.alignments = None
self.iteration = iteration
self.strict_id_match = strict_id_match # to control id match in query and target lines
#if iteration information is not given,
#yet still use_iteration_info is True,
#try to parse the profile filename
if self.iteration == None:
result_dir, basename, iter, suffix = parse_profile_filename(
result_fn)
self.iteration = iter
if self.result_fn:
self.alignments = self.parse()
def __init__(self,
queryid=None,
result_fn=None,
iteration=None,
use_iteration_info=True):
'''
COMPASS parses the COMPASS result file.
Its major functionality is parsing the alignments and
scores and return the GenericPairwiseAlignment object
So that the alignments and scores can be easily manipulated.
'''
self.queryid = queryid
self.result_fn = result_fn
self.alignments = None
self.iteration = iteration
#if iteration information is not given,
#yet still use_iteration_info is True,
#try to parse the profile filename
if self.iteration == None:
result_dir, basename, iter, suffix = parse_profile_filename(
result_fn)
self.iteration = iter
if self.result_fn:
self.alignments = self.parse()
def set_result_files( self ) :
'''
returns list of a3m files in the save_dir.
'''
basename = parse_profile_filename( self.sequence )[1]
if not basename :
return
a3m_files = glob.glob( os.path.join( self.save_dir, basename + '*.a3m' ) )
new_a3m_files = []
for a3m in a3m_files :
dir, basename2, iteration, ext = parse_profile_filename( a3m )
newa3m = build_profile_filename( dir, basename, iteration, ext )
shutil.move( a3m, newa3m )
new_a3m_files.append( newa3m )
self.a3m_files = new_a3m_files
self.a3m_files.sort()
def get_iteration_file(self, i):
'''
returns filename that contains the given iteration.
If the filename with iteration is found, it will return None.
'''
for a3m_file in self.a3m_files:
iteration = parse_profile_filename(a3m_file)[-2]
if int(i) == int(iteration):
return a3m_file
return None
def __init__(self, input_file=None, output_file=None):
if input_file == None:
raise TypeError("No Input file is given.")
if output_file == None:
dir, base_filename, iteration, ext = parse_profile_filename(
input_file)
output_file = build_profile_filename(dir, base_filename, iteration,
".pnp")
self.input_file = input_file
self.output_file = output_file
def __init__(self,
cmd=None,
input_file=None,
output_file=None,
calibration_db=None,
calibrate=True,
calibration_cmd=None):
if cmd == None:
cmd = Settings.get("hhmake")
if input_file == None:
raise TypeError("Input_file should be given.")
if calibration_cmd == None:
calibration_cmd = Settings.get("hhsearch_cmd")
if calibration_db == None:
calibration_db = Settings.get("hhm_cal_db")
if output_file == None:
dir, base_filename, iteration, ext = parse_profile_filename(
input_file)
output_file = build_profile_filename(dir, base_filename, iteration,
'.hhm')
self.cmd = cmd
self.calibration_cmd = calibration_cmd
self.calibration_db = calibration_db
self.calibrate = calibrate
self.input_file = input_file
self.output_file = output_file
#the following part is added due to hhsearch bug
#of cannot handle long input file name handling!
self.tmpinput = tempfile.NamedTemporaryFile()
self.tmpinputname = self.tmpinput.name
try:
shutil.copy(self.input_file, self.tmpinputname)
except IOError:
self.tmpinputname = self.input_file
self.tmpoutput = tempfile.NamedTemporaryFile()
self.tmpoutputname = self.tmpoutput.name
#need to be copied after the execution!
self.command_lines = self.get_command_lines()
def get_command_line(self):
if self.input_type == None:
self.input_type = 'a3m'
if self.output_type == None:
self.output_type = 'psi'
if self.output_file == None:
dir, base_filename, iteration, ext = parse_profile_filename(
self.input_file)
self.output_file = build_profile_filename(dir, base_filename,
iteration,
'.' + self.output_type)
command_line = [
self.cmd, self.input_type, self.output_type, self.input_file,
self.output_file
]
return command_line
def get_command_line(self):
if self.output_file == None:
dir, base_filename, iteration, ext = parse_profile_filename(
self.input_file)
if verbose:
print("COMPASS Builder file analysis")
print('input_file', self.input_file)
print("dir:", dir)
print("base_filename:", base_filename)
print("ieration:", iteration)
print("ext:", ext)
self.output_file = build_profile_filename(dir, base_filename,
iteration, '.cnp')
self.temporary_output_fp = tempfile.NamedTemporaryFile()
self.temporary_output_file = self.temporary_output_fp.name
list_fn = self.prepare_list_file()
command_line = [
self.cmd, '-i', list_fn, '-o', self.temporary_output_file
]
return command_line
def check_generated_profile(dominfo):
'''
returns number of iterations when all iteration results
meet integrity check criteria.
Currently, simply check
the existence of the all iteration files have the same
number of iterations.
A more rigorous checking should be implemented
e.g. checking the contents of each profile type.
'''
if verbose:
print('checking profile integrity...', dominfo['uniqueid'])
if dominfo.get('profile_integrity'):
return dominfo['profile_integrity']
domid = dominfo['uniqueid']
dompath = dominfo['domain_path']
if not os.path.exists(dompath):
return 0
basename = os.path.join(dompath, domid)
all_profiles = glob.glob(os.path.join(basename + ".*.*"))
pnp = [
fn for fn in all_profiles
if fn.endswith('.pnp') and parse_profile_filename(fn)[2]
]
a3m = [
fn for fn in all_profiles
if fn.endswith('.a3m') and parse_profile_filename(fn)[2]
]
hhm = [
fn for fn in all_profiles
if fn.endswith('.hhm') and parse_profile_filename(fn)[2]
]
cnp = [
fn for fn in all_profiles
if fn.endswith('.cnp') and parse_profile_filename(fn)[2]
]
cnplen = [
fn for fn in all_profiles
if fn.endswith('.cnp.len') and parse_profile_filename(fn)[2]
]
#checking single item
#for null case to be the True.
if not pnp:
if verbose:
print("No result file is found!")
return 0
if len(pnp) == len(a3m) == len(hhm) == len(cnp) == len(cnplen):
#need to continue to test a little more on the maximum
#iteration!!
pass
else:
return 0
try:
max_pnpi = max([int(parse_profile_filename(fn)[-2]) for fn in pnp])
max_a3mi = max([int(parse_profile_filename(fn)[-2]) for fn in a3m])
max_hhmi = max([int(parse_profile_filename(fn)[-2]) for fn in hhm])
max_cnpi = max([int(parse_profile_filename(fn)[-2]) for fn in cnp])
max_cnpleni = max(
[int(parse_profile_filename(fn)[-2]) for fn in cnplen])
if max_pnpi == max_a3mi == max_hhmi == max_cnpi == max_cnpleni == len(
pnp):
return max_pnpi
else:
return 0
except ValueError:
if verbose:
print("The filename parsing routine has a problem!")
print(pnp)
print(a3m)
print(hhm)
print(cnp)
print(cnplen)
return 0