def setup_class(cls):
cls.fpath = input_dir / 'ecSecB_apo.csv'
data = read_dynamx(cls.fpath)
control = ('Full deuteration control', 0.167*60)
cls.temperature, cls.pH = 273.15 + 30, 8.
pf = PeptideMasterTable(data, drop_first=1, ignore_prolines=True, remove_nan=False)
pf.set_control(control)
cls.hdxm = HDXMeasurement(pf.get_state('SecB WT apo'), temperature=cls.temperature, pH=cls.pH)
initial_rates = csv_to_dataframe(output_dir / 'ecSecB_guess.csv')
gibbs_guess = cls.hdxm.guess_deltaG(initial_rates['rate'])
cls.fit_result = fit_gibbs_global(cls.hdxm, gibbs_guess, epochs=100, r1=2)
from pyhdx.fitting import fit_gibbs_global_batch
from pyhdx.fitting_torch import CheckPoint
from pyhdx.models import PeptideMasterTable, HDXMeasurement, HDXMeasurementSet
current_dir = Path(__file__).parent
#current_dir = Path().cwd() / 'templates' # pycharm scientific compat
output_dir = current_dir / 'output'
output_dir.mkdir(exist_ok=True)
data_dir = current_dir.parent / 'tests' / 'test_data'
data = read_dynamx(data_dir / 'input' / 'ecSecB_apo.csv', data_dir / 'input' / 'ecSecB_dimer.csv')
pmt = PeptideMasterTable(data)
pmt.set_control(('Full deuteration control', 0.167*60))
st1 = HDXMeasurement(pmt.get_state('SecB his dimer apo'), pH=8, temperature=273.15 + 30)
st2 = HDXMeasurement(pmt.get_state('SecB WT apo'), pH=8, temperature=273.15 + 30)
hdx_set = HDXMeasurementSet([st1, st2])
guess = csv_to_protein(data_dir / 'output' / 'ecSecB_guess.csv')
gibbs_guess = hdx_set[0].guess_deltaG(guess['rate'])
# Example fit with only 5000 epochs and high learning rate
# Checkpoint stores model history every `epoch_step` epochs
checkpoint = CheckPoint(epoch_step=250)
result = fit_gibbs_global_batch(hdx_set, gibbs_guess, r1=0.5, r2=0.1, epochs=5000, lr=1e5, callbacks=[checkpoint])
print(f"MSE loss: {result.mse_loss:.2f}, "
f"Reg loss: {result.reg_loss:.2f}, "
f"Reg percent: {result.regularization_percentage:.0f}%")
def load_from_yaml(yaml_dict,
data_dir=None,
**kwargs): #name: load what from yaml?
#todo perhas classmethod on HDXMeasurement object?
"""
Creates a :class:`~pyhdx.models.HDXMeasurement` object from dictionary input.
Dictionary can be generated from .yaml format and should specify
Parameters
----------
yaml_dict : :obj:`dict`
Input dictionary specifying metadata and file location to load
data_dir : :obj:`str` or pathlib.Path object
Returns
-------
hdxm : :class:`~pyhdx.models.HDXMeasurement`
Output data object as specified by `yaml_dict`.
"""
if data_dir is not None:
input_files = [
Path(data_dir) / fname for fname in yaml_dict['filenames']
]
else:
input_files = yaml_dict['filenames']
data = read_dynamx(*input_files)
pmt = PeptideMasterTable(data, d_percentage=yaml_dict['d_percentage']
) #todo add proline, n_term options
if 'control' in yaml_dict.keys(
): # Use a FD control for back exchange correction
pmt.set_control(tuple(yaml_dict['control']))
elif 'be_percent' in yaml_dict.keys(
): # Flat back exchange percentage for all peptides\
pmt.set_backexchange(yaml_dict['be_percent'])
else:
raise ValueError('No valid back exchange control method specified')
if yaml_dict['temperature_unit'].lower() == 'celsius':
temperature = yaml_dict['temperature'] + 273.15
elif yaml_dict['temperature_unit'].lower() == 'kelvin':
temperature = yaml_dict['temperature']
else:
raise ValueError(
"Invalid option for 'temperature_unit', must be 'Celsius' or 'Kelvin'"
)
sequence = yaml_dict.get('sequence', '')
c_term = yaml_dict.get('c_term', 0)
n_term = yaml_dict.get('n_term', 1)
if not (c_term or sequence):
raise ValueError("Must specify either 'c_term' or 'sequence'")
state_data = pmt.get_state([yaml_dict['series_name']])
hdxm = HDXMeasurement(state_data,
temperature=temperature,
pH=yaml_dict['pH'],
sequence=sequence,
n_term=n_term,
c_term=c_term,
**kwargs)
return hdxm
np.random.seed(43)
data_dir = current_dir.parent / 'tests' / 'test_data'
data = read_dynamx(data_dir / 'ecSecB_apo.csv', data_dir / 'ecSecB_dimer.csv')
pmt = PeptideMasterTable(data)
pmt.set_control(('Full deuteration control', 0.167))
names = ['SecB his dimer apo', 'SecB WT apo']
data_objs = []
for name in names:
data = pmt.get_state(name)
bools = data['end'] < 40
selected_data = data[bools]
st = HDXMeasurement(selected_data)
#st = HDXMeasurement(data)
data_objs.append(st)
if __name__ == '__main__':
client = default_client()
futures = client.map(fit_rates_weighted_average,
data_objs,
client='worker_client')
print(futures)
def future_done(future):
print('joohoehoeho', future)
def load_from_yaml_v040b2(yaml_dict,
data_dir=None,
**kwargs): # pragma: no cover
"""
This is the legacy version to load yaml files of PyHDX v0.4.0b2
Creates a :class:`~pyhdx.models.HDXMeasurement` object from dictionary input.
Dictionary can be generated from .yaml format. See templates/yaml_files/SecB.yaml for format specification.
Parameters
----------
yaml_dict : :obj:`dict`
Input dictionary specifying experimental metadata and file location to load
data_dir : :obj:`str` or pathlib.Path object
Returns
-------
hdxm : :class:`~pyhdx.models.HDXMeasurement`
Output data object as specified by `yaml_dict`.
"""
if data_dir is not None:
input_files = [
Path(data_dir) / fname for fname in yaml_dict["filenames"]
]
else:
input_files = yaml_dict["filenames"]
data = read_dynamx(*input_files)
pmt = PeptideMasterTable(data, d_percentage=yaml_dict["d_percentage"]
) # todo add proline, n_term options
if "control" in yaml_dict.keys(
): # Use a FD control for back exchange correction
pmt.set_control(tuple(yaml_dict["control"]))
elif ("be_percent" in yaml_dict.keys()
): # Flat back exchange percentage for all peptides\
pmt.set_backexchange(yaml_dict["be_percent"])
else:
raise ValueError("No valid back exchange control method specified")
if yaml_dict["temperature_unit"].lower() == "celsius":
temperature = yaml_dict["temperature"] + 273.15
elif yaml_dict["temperature_unit"].lower() == "kelvin":
temperature = yaml_dict["temperature"]
else:
raise ValueError(
"Invalid option for 'temperature_unit', must be 'Celsius' or 'Kelvin'"
)
sequence = yaml_dict.get("sequence", "")
c_term = yaml_dict.get("c_term", 0)
n_term = yaml_dict.get("n_term", 1)
if not (c_term or sequence):
raise ValueError("Must specify either 'c_term' or 'sequence'")
state_data = pmt.get_state([yaml_dict["series_name"]])
hdxm = HDXMeasurement(state_data,
temperature=temperature,
pH=yaml_dict["pH"],
sequence=sequence,
n_term=n_term,
c_term=c_term,
**kwargs)
return hdxm
def yaml_to_hdxm(yaml_dict, data_dir=None, data_filters=None, **kwargs):
# todo perhas classmethod on HDXMeasurement object?
"""
Creates a :class:`~pyhdx.models.HDXMeasurement` object from dictionary input.
Dictionary can be generated from .yaml format. See templates/yaml_files/SecB.yaml for format specification.
Parameters
----------
yaml_dict : :obj:`dict`
Input dictionary specifying experimental metadata and file location to load
data_dir : :obj:`str` or pathlib.Path object
Returns
-------
hdxm : :class:`~pyhdx.models.HDXMeasurement`
Output data object as specified by `yaml_dict`.
"""
if data_dir is not None:
input_files = [
Path(data_dir) / fname for fname in yaml_dict["filenames"]
]
else:
input_files = yaml_dict["filenames"]
data = read_dynamx(*input_files)
pmt = PeptideMasterTable(data,
drop_first=yaml_dict.get('drop_first', 1),
d_percentage=yaml_dict['d_percentage']
) #todo add proline, n_term options
if 'control' in yaml_dict.keys(
): # Use a FD control for back exchange correction
# todo control should be set from an external file
control_state = yaml_dict["control"]["state"]
exposure_value = yaml_dict["control"]["exposure"]["value"]
exposure_units = yaml_dict["control"]["exposure"]["unit"]
control_exposure = exposure_value * time_factors[exposure_units]
pmt.set_control((control_state, control_exposure))
elif ("be_percent" in yaml_dict.keys()
): # Flat back exchange percentage for all peptides\
pmt.set_backexchange(yaml_dict["be_percent"])
else:
raise ValueError("No valid back exchange control method specified")
temperature = yaml_dict["temperature"]["value"]
try:
t_offset = temperature_offsets[yaml_dict["temperature"]["unit"]]
except KeyError:
t_offset = temperature_offsets[yaml_dict["temperature"]
["unit"].lower()]
temperature += t_offset
sequence = yaml_dict.get("sequence", "")
c_term = yaml_dict.get("c_term")
n_term = yaml_dict.get("n_term") or 1
if not (c_term or sequence):
raise ValueError("Must specify either 'c_term' or 'sequence'")
state_data = pmt.get_state(yaml_dict["state"])
data_filters = data_filters or []
for filter in data_filters:
state_data = filter(state_data)
hdxm = HDXMeasurement(state_data,
temperature=temperature,
pH=yaml_dict["pH"],
sequence=sequence,
n_term=n_term,
c_term=c_term,
**kwargs)
return hdxm