def look_for_pssms_in_dir(directory):
"""Find the models that are in the specified directory (named <dataset>-<cross fold index>.pssm)."""
# for each file called *.pssm in the model directory
logging.info('Looking for PSSMs in: %s' % directory)
parsed_models = dict()
for file in glob.glob(os.path.join(directory, '*-*.pssm')):
try:
#logging.info('Found models in %s' % file)
base, ext = os.path.splitext(os.path.basename(file))
dataset, cross_fold_idx = base.split('-')
cross_fold_idx = int(cross_fold_idx)-1
logging.info(
'Read models from: %s for dataset: %s and cross-fold validation set: %d',
file,
dataset,
cross_fold_idx
)
parsed_models[(dataset, cross_fold_idx)] = list(
parse_models(
open(os.path.join(file))
)
)
except:
print sys.exc_info()
logging.warning('Could not parse: %s' % file)
return parsed_models
def __init__(self, pssm_file, log):
self.pssm_file = pssm_file
self.png_file = pssm_file.replace('.pssm', '.png')
self.eps_file = pssm_file.replace('.pssm', '.eps')
re_match = Pssm.pssm_file_name_re.match(os.path.basename(pssm_file))
self.tag = re_match.group(1)
self.pssm_idx = int(re_match.group(2))
self.num_sites, self.num_seqs_with_site, self.num_seqs = log.site_numbers[self.pssm_idx]
# logging.info('%s: %s %d %d', self.pssm_file, self.fragment, self.cross_fold, self.pssm_idx)
self.model = parse_models(open(self.pssm_file)).next()
self.emissions = calculate_emissions(self.model)
self.gap_probs = calculate_gap_probs(self.model)
self.first_order_entropy_score = calculate_first_order_entropy_score(self.emissions)
self.information_content_score = calculate_information_content_score(self.emissions)
self.num_seqs_with_site_score = float(self.num_seqs_with_site) / float(self.num_seqs)
self.overall_score = weighted_geometric_mean(
(self.first_order_entropy_score, self.information_content_score, self.num_seqs_with_site_score),
[1.5 , 1. , 1.]
)
logging.info(
'%s; %8g; %8g; %8g; %8g',
self.pssm_file,
self.first_order_entropy_score,
self.information_content_score,
self.num_seqs_with_site_score,
self.overall_score
)
def look_for_pssms_in_dir(directory):
"""Find the models that are in the specified directory (named <dataset>-<cross fold index>.pssm)."""
# for each file called *.pssm in the model directory
logging.info('Looking for PSSMs in: %s' % directory)
parsed_models = dict()
for file in glob.glob(os.path.join(directory, '*-*.pssm')):
try:
#logging.info('Found models in %s' % file)
base, ext = os.path.splitext(os.path.basename(file))
dataset, cross_fold_idx = base.split('-')
cross_fold_idx = int(cross_fold_idx) - 1
logging.info(
'Read models from: %s for dataset: %s and cross-fold validation set: %d',
file, dataset, cross_fold_idx)
parsed_models[(dataset, cross_fold_idx)] = list(
parse_models(open(os.path.join(file))))
except:
print sys.exc_info()
logging.warning('Could not parse: %s' % file)
return parsed_models
help="File in which the gapped PSSMs are stored.")
option_parser.add_option(
"-l",
"--logo-files-basename",
dest="logo_files_basename",
help="basename of files to write logos to. Extension will be -0.png")
option_parser.add_option("-t",
"--image-type",
dest="image_type",
default='png',
help="type of images to write")
options, args = option_parser.parse_args()
for option in option_parser.option_list:
if option.dest:
logging.info('%s: %s (%s)', option.dest,
str(getattr(options, option.dest)), option.help)
# Load PSSMs
logging.info('Loading PSSMs: %s', options.models_file)
pssms = list(parse_models(open(options.models_file)))
for i, p in enumerate(pssms):
filename = '%s-%d.%s' % (options.logo_files_basename, i,
options.image_type)
logging.info('Creating image for PSSM: %s', filename)
emissions, gap_probs = emissions_and_gaps_from_semi_parsed(p)
logo_image = logo.pssm_as_image(emissions, transparencies=gap_probs)
logo_image.save(filename)
#model_dir = '/home/reid/Analysis/GappedPssms/apr-2009/single-gap'
#sequence_dir = '/home/reid/Data/GappedPssms/apr-2009/'
fisher_p_values = list()
for fragment, pssm in pssms():
sequence_file = os.path.join(sequence_dir,
sequence_filename_fmt % fragment)
model_file = os.path.join(model_dir, '%s-%s.pssm' % (fragment, pssm))
logging.info('Loading sequences: %s', sequence_file)
sequences = list(sequences_from_fasta(sequence_file))
numpy_seqs = map(seq_to_numpy, sequences)
logging.info('Loaded %d sequences', len(sequences))
logging.info('Parsing PSSMs: %s', model_file)
pssms = list(parse_models(open(model_file)))
logging.info('Building models')
models = [
build_hmm_from_semi_parsed(parsed, p_binding_site=p_binding_site)
for parsed in pssms
]
def nucleotide_dist():
return numpy.zeros(4) + .25
base_dists = DictOf(nucleotide_dist)
min_site_length = 20
logging.info('Analysing sequences')
for hmm, traits in models:
help="basename of files to write logos to. Extension will be -0.png"
)
option_parser.add_option(
"-t",
"--image-type",
dest="image_type",
default='png',
help="type of images to write"
)
options, args = option_parser.parse_args()
for option in option_parser.option_list:
if option.dest:
logging.info('%s: %s (%s)', option.dest, str(getattr(options, option.dest)), option.help)
# Load PSSMs
logging.info('Loading PSSMs: %s', options.models_file)
pssms = list(parse_models(open(options.models_file)))
for i, p in enumerate(pssms):
filename = '%s-%d.%s' % (options.logo_files_basename, i, options.image_type)
logging.info('Creating image for PSSM: %s', filename)
emissions, gap_probs = emissions_and_gaps_from_semi_parsed(p)
logo_image = logo.pssm_as_image(
emissions,
transparencies=gap_probs
)
logo_image.save(filename)
#sequence_dir = '/home/reid/Data/GappedPssms/apr-2009/'
fisher_p_values = list()
for fragment, pssm in pssms():
sequence_file = os.path.join(sequence_dir, sequence_filename_fmt % fragment)
model_file = os.path.join(model_dir, '%s-%s.pssm' % (fragment, pssm))
logging.info('Loading sequences: %s', sequence_file)
sequences = list(sequences_from_fasta(sequence_file))
numpy_seqs = map(seq_to_numpy, sequences)
logging.info('Loaded %d sequences', len(sequences))
logging.info('Parsing PSSMs: %s', model_file)
pssms = list(parse_models(open(model_file)))
logging.info('Building models')
models = [
build_hmm_from_semi_parsed(parsed, p_binding_site=p_binding_site)
for parsed in pssms
]
def nucleotide_dist():
return numpy.zeros(4) + .25
base_dists = DictOf(nucleotide_dist)
min_site_length = 20
logging.info('Analysing sequences')
for hmm, traits in models:
logging.info('Looking for PSSMs in: %s' % options.model_dir)
pssms_for_method = dict() # a dict of parsed models indexed by dataset/cross-validation set index tuple
for file in os.listdir(options.model_dir):
base, ext = os.path.splitext(file)
if '.pssm' == ext:
dataset, cross_fold_idx = base.split('-')
cross_fold_idx = int(cross_fold_idx)
logging.info(
'Reading models from: %s for dataset: %s and cross-fold validation set: %d',
file,
dataset,
cross_fold_idx
)
pssms_for_method[(dataset, cross_fold_idx)] = list(
parse_models(
open(os.path.join(options.model_dir, file))
)
)
#
# Generate a ROC point for each p_binding_site for these PSSMs
#
#p_binding_sites = [.1]
num_roc_points = int(options.num_points)
min_p_binding_site = float(options.max_p_binding_site)
max_p_binding_site = float(options.min_p_binding_site)