names=["q", "SA", "sigSA"],
delim_whitespace=True,
skiprows=1)
### correct the average-off protein-only curve for packing effects
spf_corrected_int = avg_off.SA / spf.SA
spf_corrected_error = spf_corrected_int * np.sqrt(
(avg_off.sigSA / avg_off.SA)**2 + (spf.sigSA / spf.SA)**
2) ### https://terpconnect.umd.edu/~toh/models/ErrorPropagation.pdf
avg_off_spf_corrected = Trace(avg_off.q, np.empty_like(avg_off.q),
np.empty_like(avg_off.q), spf_corrected_error,
spf_corrected_int, np.empty_like(avg_off.q))
### scale back up to proper intensities
avg_off_spf_corrected.scale(avg_off,
qmin=0.03,
qmax=0.1,
approach="projection")
avg_off_spf_corrected_scaled = Trace(avg_off.q, np.empty_like(avg_off.q),
np.empty_like(avg_off.q),
avg_off_spf_corrected.scaled_sigSA,
avg_off_spf_corrected.scaled_SA,
np.empty_like(avg_off.q))
avg_off_spf_corrected_scaled.write_dat(
avg_off_protein_only_file.replace('.dat', 'spf-corrected.dat'))
# plt.plot(avg_off.q,avg_off.SA, label='avg_off')
# plt.plot(avg_off_spf_corrected.q,avg_off_spf_corrected.SA, label='avg_off_spf_corr')
# plt.plot(avg_off_spf_corrected_scaled.q,avg_off_spf_corrected_scaled.SA, label='avg_off_spf_corr_scaled')
### add corrected protein-only signal to each TR difference signal
for file in tr_diff_files:
name = file.name
def all_vectors(parent,
samp,
reps,
on_off_map,
option=None,
multitemp=None,
iterations=None,
temp=None):
"""
Collect all curves within a directory, selecting only on-off pairs,
and sorting into order of time-resolved data collection.
This is useful for SVD analysis.
Parameters:
parent (pathlib object): the directory where files are stored
samp (str): the name of the sample, as recorded in the filename
reps (list of strings): values denoting repetition number
on_off_map (dictionary): guide to on-off pairs
option (T/F): filename idiosynchroncy, "on" label included in filename for on-off pairs
multitemp (T/F): data storage idiosynchroncy, multiple temperatures collected in one run
iterations (T/F): data collection / storage idiosynchroncy
temp (list of strings): values denoting collection temperature
Returns:
all_vectors (list of Trace objects)
all_labels (list of strings)
tr_vectors_labels (list of tuples containing (np.array, str))
"""
all_vectors = []
all_labels = []
tr_vectors_labels = []
if args.multitemp:
for iteration in iterations:
for n in reps:
# for off in on_off_map.values():
for on, off in on_off_map.items():
on_string = ("{0}/{1}_{2}_{3}_{4}_{5}.tpkl".format(
parent, samp, iteration, temp, n, on))
off_string = ("{0}/{1}_{2}_{3}_{4}_{5}.tpkl".format(
parent, samp, iteration, temp, n, off))
try:
on_data = parse.parse(on_string)
on_data = Trace(on_data.q, np.empty_like(on_data.q),
np.empty_like(on_data.q),
on_data.sigSA, on_data.SA, on_data.Nj)
on_data.scale(reference, qmin=QMIN, qmax=QMAX)
off_data = parse.parse(off_string)
off_data = Trace(off_data.q, np.empty_like(off_data.q),
np.empty_like(off_data.q),
off_data.sigSA, off_data.SA,
off_data.Nj)
off_data.scale(reference, qmin=QMIN, qmax=QMAX)
# off_scaled = Trace(off_data.q, np.empty_like(off_data.q), np.empty_like(off_data.q), off_data.scaled_sigSA, off_data.scaled_SA, off_data.Nj)
# off_vectors.append(off_scaled)
if on_data:
if off_data:
all_vectors.append(
off_data.SA[reference.q > 0.03])
all_labels.append(off)
all_vectors.append(
on_data.SA[reference.q > 0.03])
all_labels.append(on)
# sub_scaled = subtract_scaled_traces(on_data,off_data)
sub_scaled = on_data.subtract(off_data,
scaled=True)
tr_vectors_labels.append(
(sub_scaled.SA[reference.q > 0.03], on))
except:
print(off_string + "\tfailed")
else:
for n in reps:
# for off in on_off_map.values():
for on, off in on_off_map.items():
on_string = ("{0}/{1}_{2}_{3}.tpkl".format(
parent, samp, n, on))
off_string = ("{0}/{1}_{2}_{3}.tpkl".format(
parent, samp, n, off))
try:
on_data = parse.parse(on_string)
on_data = Trace(on_data.q, np.empty_like(on_data.q),
np.empty_like(on_data.q), on_data.sigSA,
on_data.SA, on_data.Nj)
on_data.scale(reference, qmin=QMIN, qmax=QMAX)
off_data = parse.parse(off_string)
off_data = Trace(off_data.q, np.empty_like(off_data.q),
np.empty_like(off_data.q), off_data.sigSA,
off_data.SA, off_data.Nj)
off_data.scale(reference, qmin=QMIN, qmax=QMAX)
# off_scaled = Trace(off_data.q, np.empty_like(off_data.q), np.empty_like(off_data.q), off_data.scaled_sigSA, off_data.scaled_SA, off_data.Nj)
# off_vectors.append(off_scaled)
if on_data:
if off_data:
all_vectors.append(off_data.SA[reference.q > 0.03])
all_labels.append(off)
all_vectors.append(on_data.SA[reference.q > 0.03])
all_labels.append(on)
# sub_scaled = subtract_scaled_traces(on_data,off_data)
sub_scaled = on_data.subtract(off_data,
scaled=True)
tr_vectors_labels.append(
(sub_scaled.SA[reference.q > 0.03], on))
except:
print(off_string + "\tfailed")
return all_vectors, all_labels, tr_vectors_labels
delim_whitespace=True,
skiprows=1)
for file in static_files:
name = file.name
orig = parse(name)
### correct the curve for packing effects
spf_corrected_int = orig.SA / spf.SA
spf_corrected_error = spf_corrected_int * np.sqrt(
(orig.sigSA / orig.SA)**2 + (spf.sigSA / spf.SA)**
2) ### https://terpconnect.umd.edu/~toh/models/ErrorPropagation.pdf
static_spf_corrected = Trace(orig.q, np.empty_like(orig.q),
np.empty_like(orig.q), spf_corrected_error,
spf_corrected_int, np.empty_like(orig.q))
# ### scale back up to proper intensities
static_spf_corrected.scale(orig,
qmin=0.03,
qmax=0.1,
approach="projection")
static_spf_corrected_scaled = Trace(orig.q, np.empty_like(orig.q),
np.empty_like(orig.q),
static_spf_corrected.scaled_sigSA,
static_spf_corrected.scaled_SA,
np.empty_like(orig.q))
static_spf_corrected_scaled.write_dat(
name.replace('.dat', 'spf-corrected.dat'))
# plt.xscale('log')
# plt.legend()
# plt.show()
def all_vectors(parent,
samp,
reps,
on_off_map,
option=None,
multitemp=None,
iterations=None,
temp=None):
all_vectors = []
all_labels = []
tr_vectors_labels = []
if args.multitemp:
for iteration in iterations:
for n in reps:
# for off in on_off_map.values():
for on, off in on_off_map.items():
on_string = ("{0}/{1}_{2}_{3}_{4}_{5}.tpkl".format(
parent, samp, iteration, temp, n, on))
off_string = ("{0}/{1}_{2}_{3}_{4}_{5}.tpkl".format(
parent, samp, iteration, temp, n, off))
try:
on_data = parse.parse(on_string)
on_data = Trace(on_data.q, np.empty_like(on_data.q),
np.empty_like(on_data.q),
on_data.sigSA, on_data.SA, on_data.Nj)
on_data.scale(reference, qmin=QMIN, qmax=QMAX)
off_data = parse.parse(off_string)
off_data = Trace(off_data.q, np.empty_like(off_data.q),
np.empty_like(off_data.q),
off_data.sigSA, off_data.SA,
off_data.Nj)
off_data.scale(reference, qmin=QMIN, qmax=QMAX)
# off_scaled = Trace(off_data.q, np.empty_like(off_data.q), np.empty_like(off_data.q), off_data.scaled_sigSA, off_data.scaled_SA, off_data.Nj)
# off_vectors.append(off_scaled)
if on_data:
if off_data:
all_vectors.append(
off_data.SA[reference.q > 0.03])
all_labels.append(off)
all_vectors.append(
on_data.SA[reference.q > 0.03])
all_labels.append(on)
# sub_scaled = subtract_scaled_traces(on_data,off_data)
sub_scaled = on_data.subtract(off_data,
scaled=True)
tr_vectors_labels.append(
(sub_scaled.SA[reference.q > 0.03], on))
except:
print(off_string + "\tfailed")
else:
for n in reps:
# for off in on_off_map.values():
for on, off in on_off_map.items():
on_string = ("{0}/{1}_{2}_{3}.tpkl".format(
parent, samp, n, on))
off_string = ("{0}/{1}_{2}_{3}.tpkl".format(
parent, samp, n, off))
try:
on_data = parse.parse(on_string)
on_data = Trace(on_data.q, np.empty_like(on_data.q),
np.empty_like(on_data.q), on_data.sigSA,
on_data.SA, on_data.Nj)
on_data.scale(reference, qmin=QMIN, qmax=QMAX)
off_data = parse.parse(off_string)
off_data = Trace(off_data.q, np.empty_like(off_data.q),
np.empty_like(off_data.q), off_data.sigSA,
off_data.SA, off_data.Nj)
off_data.scale(reference, qmin=QMIN, qmax=QMAX)
# off_scaled = Trace(off_data.q, np.empty_like(off_data.q), np.empty_like(off_data.q), off_data.scaled_sigSA, off_data.scaled_SA, off_data.Nj)
# off_vectors.append(off_scaled)
if on_data:
if off_data:
all_vectors.append(off_data.SA[reference.q > 0.03])
all_labels.append(off)
all_vectors.append(on_data.SA[reference.q > 0.03])
all_labels.append(on)
# sub_scaled = subtract_scaled_traces(on_data,off_data)
sub_scaled = on_data.subtract(off_data,
scaled=True)
tr_vectors_labels.append(
(sub_scaled.SA[reference.q > 0.03], on))
except:
print(off_string + "\tfailed")
return all_vectors, all_labels, tr_vectors_labels
