def _args():
import argparse
parser = argparse.ArgumentParser()
parser.add_argument('--trained_model_directory')
parser.add_argument('--depth_file_name')
parser.add_argument('--chromosome_number', type=int)
parser.add_argument('--start', type=int)
parser.add_argument('--end', type=int)
parser.add_argument('--fold_reduction_of_sample_size', type=float)
parser.add_argument('--window_half_width', type=int)
parser.add_argument('--resampling_target_file_name')
parser.add_argument('--filter')
parser.add_argument('--batch_size', type=int)
parser.add_argument('--learning_rate', type=float)
args = parser.parse_args()
args.fasta_file = '../data/sequences/human_g1k_v37.fasta'
args_to_save = args.__dict__.copy()
# this allows us to pass "None" to bash script to indicate that there is no json file describing a resampling scheme
if args.resampling_target_file_name == "None":
args_to_save['resampling_target'] = "None"
args.resampling_target = None
else:
with open(args.resampling_target_file_name, 'r') as fp:
args_to_save['resampling_target'] = json.load(fp)
args.resampling_target = args_to_save['resampling_target'].copy()
args.resampling_target['function'] = getattr(
load_preprocess_data, args.resampling_target['function'])
args.resampling_target = named_tuple(args.resampling_target)
args.filter = getattr(load_preprocess_data, args.filter)
return named_tuple(args.__dict__), args_to_save
def _compute_observed_depth_mean(chromosome_number, depth_file_name):
from utility import named_tuple
from load_preprocess_data import read_depths
depths = read_depths(named_tuple({'chromosome_number': chromosome_number,
'depth_file_name': depth_file_name}))
from load_preprocess_data import compute_observed_depth_mean
return compute_observed_depth_mean(depths, chromosome_number)
def _args():
import argparse
parser = argparse.ArgumentParser()
parser.add_argument('--depth_file_name')
parser.add_argument('--trained_model_directory')
parser.add_argument('--test_directory')
parser.add_argument('--filter')
parser.add_argument('--chromosome_number', type=int)
parser.add_argument('--content_start', type=int)
parser.add_argument('--content_end', type=int)
parser.add_argument('--number_test_examples', type=int)
parser.add_argument('--padding', type=float)
args = parser.parse_args()
# high-quality deletion on chromosome 1
# region_start = (189704000 - 100000)
# region_end = (189783300 + 100000)
_compute_start_end(args)
from utility import get_train_args
args.window_half_width = get_train_args(args.trained_model_directory)['window_half_width']
args.fasta_file = '../data/sequences/human_g1k_v37.fasta'
_dump_json(args)
import load_preprocess_data
args.filter = getattr(load_preprocess_data, args.filter)
from utility import named_tuple
return named_tuple(args.__dict__)
def get_resampling_target(trained_model_directory):
with open(
get_train_args(trained_model_directory)
['resampling_target_file_name'], 'r') as f:
dictionary = json.load(f)
dictionary['function'] = getattr(load_preprocess_data,
dictionary['function'])
print(dictionary['function'])
return named_tuple(dictionary)
def plot_corrected_depths_test_all(trained_models,
normalized_depths=True, corrected_depths=True,
chromosome_number=1, min_y=0, max_y=2,
show_title=True, grid=True):
for trained_model in trained_models:
if 'depth_file_name' in trained_model:
from utility import named_tuple
from load_preprocess_data import read_depths
depths = read_depths(named_tuple({'chromosome_number': chromosome_number,
'depth_file_name': trained_model['depth_file_name']}))
from load_preprocess_data import compute_observed_depth_mean
observed_depth_mean = compute_observed_depth_mean(depths, chromosome_number)
else:
train_sampled_data, _, _ = utility_train.unpickle(trained_model['path'])
observed_depth_mean = _compute_observed_depth_mean(train_sampled_data)
test_data = pd.read_pickle(os.path.join(trained_model['path'], trained_model['test_file_name']))
title = trained_model['annotation'] if show_title else ''
figure_file_name = trained_model['figure_file_name'] if 'figure_file_name' in trained_model else None
_plot_corrected_depths(test_data, observed_depth_mean, chromosome_number,
title=title, min_y=min_y, max_y=max_y,
normalized_depths=normalized_depths, corrected_depths=corrected_depths,
figure_file_name=figure_file_name,
grid=grid)