def test_change_protein_data_dir():
with enter_temp_directory():
create_fake_series()
yaml_file ={}
yaml_file["base_dir"] = "./fake_series"
protein = "fake_kinase1"
with enter_protein_data_dir(yaml_file, protein):
current_folder_path, current_folder_name = os.path.split(os.getcwd())
assert current_folder_name == "fake_kinase1"
return
def test_change_protein_data_dir():
with enter_temp_directory():
create_fake_series()
yaml_file = {}
yaml_file["base_dir"] = "./fake_series"
protein = "fake_kinase1"
with enter_protein_data_dir(yaml_file, protein):
current_folder_path, current_folder_name = os.path.split(
os.getcwd())
assert current_folder_name == "fake_kinase1"
return
def test_series_slicer(yaml_file, folder_name="sliced_feature_dir"):
yaml_file = load_yaml_file(yaml_file)
df_dict={}
for protein in yaml_file["protein_list"]:
with enter_protein_data_dir(yaml_file, protein):
df_dict[protein] = verboseload(os.path.join(os.getcwd(),
folder_name,"feature_descriptor.h5"))
for ind,protein in enumerate(yaml_file["protein_list"]):
for ind2, protein2 in enumerate(yaml_file["protein_list"]):
assert (df_dict[protein].resnames==
df_dict[protein2].resnames).all()
return
def test_series_slicer(yaml_file, folder_name="sliced_feature_dir"):
yaml_file = load_yaml_file(yaml_file)
df_dict={}
for protein in yaml_file["protein_list"]:
with enter_protein_data_dir(yaml_file, protein):
df_dict[protein] = verboseload(os.path.join(os.getcwd(),
folder_name,"feature_descriptor.h5"))
for ind,protein in enumerate(yaml_file["protein_list"]):
for ind2, protein2 in enumerate(yaml_file["protein_list"]):
assert (df_dict[protein].resnames==
df_dict[protein2].resnames).all()
return
def pull_features(yaml_file, prt, skip=1, feature_indices=None):
"""
Simple utility to pull certain features from the feature_folder object
:param prt: Protein model to use
:param skip: skip for each file(defaults to 1)
:param feature_indices: which indices to pull
:return: dictionary keyed on file name with feature values as arrays
"""
yaml_file = load_yaml_file(yaml_file)
all_f ={}
with enter_protein_data_dir(yaml_file, prt.name):
feature_file_list = glob.glob("./%s/*.jl"%yaml_file["feature_dir"])
for i in feature_file_list:
all_f[os.path.basename(i)]=load(i)[:, feature_indices]
return all_f
def subsample_protein(yaml_file, protein, stride=5,out_dir="sub_protein_traj"):
yaml_file=load_yaml_file(yaml_file)
p=Pool(int(cpu_count()/2))
with enter_protein_data_dir(yaml_file, protein):
flist = [os.path.abspath(i) for i in
glob.glob("%s/*.hdf5"%yaml_file["protein_dir"])]
base_dir = yaml_file["base_dir"]
new_output_dir = os.path.join(base_dir,protein,out_dir)
if not os.path.isdir(new_output_dir):
os.mkdir(new_output_dir)
fout = [os.path.join(new_output_dir,os.path.basename(i)) for i in flist]
zippy = zip(flist, fout, itertools.repeat(stride))
jobs= [(i,o,s) for i,o,s in zippy]
p.map(subsample_traj,jobs)
return
def test_slicer():
with enter_temp_directory():
base_dir = os.path.abspath(os.path.curdir)
mdl_dir = os.path.join(base_dir,"mdl_dir")
feature_dir = "feature_dir"
series_name = "fake_series"
protein_list = ["kinase_1", "kinase_2"]
project_dict = {"kinase_1": ["fake_proj1",],
"kinase_2": ["fake_proj2"]}
mdl_params = {'tica__n_components': 1, 'tica__lag_time': 1,
'tica__kinetic_mapping': True, 'tica__shrinkage': 0.01,
'cluster__n_clusters': 2,
'msm__lag_time': 1, 'bootstrap__n_samples':1 }
create_fake_data(base_dir, protein_list, project_dict)
yaml_file = setup_series_analysis(base_dir, mdl_dir, feature_dir,
series_name, protein_list,
project_dict, mdl_params)
dict_feat_ind={}
dict_feat_ind["kinase_1"] =[0, 2]
dict_feat_ind["kinase_2"] =[1, 1, 0, 2]
series_feature_slicer(yaml_file, dict_feat_ind)
for protein in protein_list:
with enter_protein_data_dir(yaml_file, protein):
assert (os.path.isdir("sliced_feature_dir"))
flist = glob.glob("./%s/*.jl"%feature_dir)
for fname in flist:
original_file = verboseload(fname)
expected_file = original_file[:, dict_feat_ind[protein]]
written_file = verboseload("./%s/%s"%("sliced_feature_dir",
os.path.basename(fname)
))
assert (expected_file==written_file).all()
return
def _get_common_features(yaml_file, featurizer, aligned_dict,save_df=True):
"""
Function to get the common features across protein using the common residues.
can optionally save the pandas data to the mdl_dir
:param yaml_file: The protein yaml_file
:param featurizer: featurizer object used.
:param prt_mapping: Mapping of each residue to its sequence
:param aligned_dict : Dictionary of alignments for each protein
:return:
"""
result_dict = {}
df_dict={}
for protein in yaml_file["protein_list"]:
print(protein)
#reset the featurizer
featurizer = clone(featurizer)
trj = load_random_traj(yaml_file, protein)
df = pd.DataFrame(featurizer.describe_features(trj))
prt_mapping, prt_seq = _map_residue_ind_seq_ind(yaml_file, protein,
aligned_dict[protein])
feature_vec =[]
#for every feature
for i in df.iterrows():
#get the index and the feature itself
feature_ind, feature_dict = i
all_res_in_algn = []
mapped_index_list=[]
for aa_ind in feature_dict["resids"]:
aa_code = prt_seq[aa_ind]
#make sure we have the same residue
assert(trj.top.residue(aa_ind).code==aa_code)
#get the mapping for that aa to the main alignment
mapped_index = prt_mapping[aa_ind]
#for every protein in the alignment, check if we have the same residue
#at the same position
all_res_in_algn.append(np.alltrue([aligned_dict[prt][mapped_index]==aa_code
for prt in yaml_file["protein_list"]]))
mapped_index_list.append(mapped_index)
#to account for additions and deletions, we check if the difference between
#the mapping and the actual residue codes is the same.
mapped_index_difference = [x - mapped_index_list[i - 1]
for i, x in enumerate(mapped_index_list) if i > 0]
resid_index_difference = [x - feature_dict["resids"][i - 1]
for i, x in enumerate(feature_dict["resids"]) if i > 0]
if not np.all(mapped_index_difference==resid_index_difference):
all_res_in_algn.append(False)
if np.alltrue(all_res_in_algn):
feature_vec.append(feature_ind)
df_dict[protein] = df.iloc[feature_vec]
result_dict[protein] = feature_vec
if save_df:
new_df = df.iloc[feature_vec]
with enter_protein_mdl_dir(yaml_file, protein):
verbosedump(new_df, os.path.join("feature_descriptor.h5"))
with enter_protein_data_dir(yaml_file, protein):
verbosedump(new_df, os.path.join("sliced_feature_dir",
"feature_descriptor.h5"))
return result_dict, df_dict
def _get_common_features(yaml_file, featurizer, aligned_dict,save_df=True):
"""
Function to get the common features across protein using the common residues.
can optionally save the pandas data to the mdl_dir
:param yaml_file: The protein yaml_file
:param featurizer: featurizer object used.
:param prt_mapping: Mapping of each residue to its sequence
:param aligned_dict : Dictionary of alignments for each protein
:return:
"""
result_dict = {}
df_dict={}
for protein in yaml_file["protein_list"]:
print(protein)
#reset the featurizer
featurizer = clone(featurizer)
trj = load_random_traj(yaml_file, protein)
df = pd.DataFrame(featurizer.describe_features(trj))
prt_mapping, prt_seq = _map_residue_ind_seq_ind(yaml_file, protein,
aligned_dict[protein])
feature_vec =[]
#for every feature
for i in df.iterrows():
#get the index and the feature itself
feature_ind, feature_dict = i
all_res_in_algn = []
mapped_index_list=[]
for aa_ind in feature_dict["resids"]:
aa_code = prt_seq[aa_ind]
#make sure we have the same residue
assert(trj.top.residue(aa_ind).code==aa_code)
#get the mapping for that aa to the main alignment
mapped_index = prt_mapping[aa_ind]
#for every protein in the alignment, check if we have the same residue
#at the same position
all_res_in_algn.append(np.alltrue([aligned_dict[prt][mapped_index]==aa_code
for prt in yaml_file["protein_list"]]))
mapped_index_list.append(mapped_index)
#to account for additions and deletions, we check if the difference between
#the mapping and the actual residue codes is the same.
mapped_index_difference = [x - mapped_index_list[i - 1]
for i, x in enumerate(mapped_index_list) if i > 0]
resid_index_difference = [x - feature_dict["resids"][i - 1]
for i, x in enumerate(feature_dict["resids"]) if i > 0]
if not np.all(mapped_index_difference==resid_index_difference):
all_res_in_algn.append(False)
if np.alltrue(all_res_in_algn):
feature_vec.append(feature_ind)
df_dict[protein] = df.iloc[feature_vec]
result_dict[protein] = feature_vec
if save_df:
new_df = df.iloc[feature_vec]
with enter_protein_mdl_dir(yaml_file, protein):
verbosedump(new_df, os.path.join("feature_descriptor.h5"))
with enter_protein_data_dir(yaml_file, protein):
verbosedump(new_df, os.path.join("sliced_feature_dir",
"feature_descriptor.h5"))
return result_dict, df_dict
