def test_apod_gmb(self):
spec = fromBruker(self.filename, False, False)
spec = apod(spec, w=lambda s: GMB(s, 2, 0.5))
C = converter()
u = guess_udic(self.dic,self.data)
C.from_bruker(self.dic, self.data, u)
pipe_dic, pipe_data = C.to_pipe()
test_data = pipep.gmb(pipe_dic, pipe_data, lb=2, gb=0.5)[1]
ts.assert_allclose(spec, test_data, 1e-7,1e-3, 'GBM apodization not equal to NMRPipe processed one.')
def test_apod_gm(self):
spec = fromBruker(self.filename, False, False)
spec = apod(spec, w=lambda s: GM(s, 1, 10, 5))
C = converter()
u = guess_udic(self.dic,self.data)
C.from_bruker(self.dic, self.data, u)
pipe_dic, pipe_data = C.to_pipe()
test_data = pipep.gm(pipe_dic, pipe_data, g1=1, g2=10, g3=5)[1]
ts.assert_allclose(spec, test_data, 1e-7,1e-3, 'GM apodization not equal to NMRPipe processed one.')
def test_apod_sp(self):
spec = fromBruker(self.filename, False, False)
spec = apod(spec, w=lambda s: SP(s, off=0.5, power=2))
C = converter()
u = guess_udic(self.dic,self.data)
C.from_bruker(self.dic, self.data, u)
pipe_dic, pipe_data = C.to_pipe()
test_data = pipep.sp(pipe_dic, pipe_data, off=0.5, pow=2)[1]
ts.assert_allclose(spec, test_data, 1e-7,1e-3, 'SP apodization not equal to NMRPipe processed one.')
def test_apod_jmod(self):
spec = fromBruker(self.filename, False, False)
spec = apod(spec, w=lambda s: JMOD(s, 0.5, 5, 1))
C = converter()
u = guess_udic(self.dic,self.data)
C.from_bruker(self.dic, self.data, u)
pipe_dic, pipe_data = C.to_pipe()
test_data = pipep.jmod(pipe_dic, pipe_data, off=0.5, j=5, lb=1)[1]
ts.assert_allclose(spec, test_data, 1e-7,1e-3, 'JMOD apodization not equal to NMRPipe processed one.')
def test_apod_tri(self):
spec = fromBruker(self.filename, False, False)
# Loc < 1 results in dimension reduction of the apodization vector
with self.assertRaises(NMRShapeError):
apod(spec, w=lambda s: TRI(s, 0.5, 0.7, 0.5))
spec = fromBruker(self.filename, False, False)
spec = apod(spec, w=lambda s: TRI(s, 5000, 0.7, 0.5))
C = converter()
u = guess_udic(self.dic,self.data)
C.from_bruker(self.dic, self.data, u)
pipe_dic, pipe_data = C.to_pipe()
test_data = pipep.tri(pipe_dic, pipe_data, loc=5000, lHi=0.7, rHi=0.5)[1]
ts.assert_allclose(spec, test_data, 1e-7,1e-3, 'TRI apodization not equal to NMRPipe processed one.')
def read_varian_spec_raw(spectrum_directory,
fid_filename,
procpar_filename,
zero_filling=True,
apodization=True):
from nmrglue.fileio import varian, convert
from nmrglue.process import pipe_proc
#Read varian files:
res = varian.read(dir=spectrum_directory,
fid_file=fid_filename,
procpar_file=procpar_filename)
varian_data = res[1]
varian_dic = res[0]
#Get main parameters for ppm scale:
sw = float(varian_dic["procpar"]["sw"]["values"][0])
obs = float(varian_dic["procpar"]["sfrq"]["values"][0])
car = float(varian_dic["procpar"]["reffrq"]["values"][0])
#Convert varian to pipe:
universal_varian_dic = varian.guess_udic(varian_dic, varian_data)
universal_varian_dic[0]["sw"] = sw
universal_varian_dic[0]["obs"] = obs
universal_varian_dic[0]["label"] = varian_dic["procpar"]["tn"]["values"][0]
universal_varian_dic[0]["car"] = car
C = convert.converter()
C.from_varian(varian_dic, varian_data, universal_varian_dic)
pipe_dic, varian_pipe_data = C.to_pipe()
pipe_dic["FDF1SW"] = sw
pipe_dic["FDF1OBS"] = obs
pipe_dic["FDF1LABEL"] = universal_varian_dic[0]["label"]
pipe_dic["FDF1CAR"] = car
#If zero filling:
if zero_filling:
pipe_dic, varian_data = pipe_proc.zf(pipe_dic, varian_data)
#If apodization:
if apodization:
lb_d = float(varian_dic["procpar"]["lb"]["values"][0])
pipe_dic, varian_data = pipe_proc.em(pipe_dic, varian_data, lb=lb_d)
#Fourier Transform:
pipe_dic, varian_data = pipe_proc.ft(pipe_dic,
varian_data,
auto=True,
inv=True)
#Phase correction:
p_zero = float(varian_dic["procpar"]["rp"]["values"][0])
p_one = float(varian_dic["procpar"]["lp"]["values"][0])
pipe_dic, varian_data = pipe_proc.ps(pipe_dic,
varian_data,
p0=p_zero,
p1=p_one,
inv=True)
#Baseline Correction:
pipe_dic, varian_data = pipe_proc.cbf(pipe_dic, varian_data)
#Remove imaginary numbers:
dic, data = pipe_proc.di(pipe_dic, varian_data)
#Calculate ppm scale:
ppm = list()
ppm_f = (obs * 1000000 + sw / 2 - (car * 1000000)) / car
ppm_width = sw / car
ppm_i = ppm_f - ppm_width
n = int(len(data))
ppm_step = ppm_width / n
ppm.append(ppm_i)
for i in range(1, n):
ppm.append(ppm[i - 1] + ppm_step)
return (ppm, abs(data))
def read_varian_spec2d_raw(spectrum_directory,
fid_filename,
procpar_filename,
zero_filling=True,
apodization=True):
from nmrglue.fileio import varian, convert
from nmrglue.process import pipe_proc
# Read varian files
res = varian.read(dir=spectrum_directory,
fid_file=fid_filename,
procpar_file=procpar_filename)
varian_data = res[1]
varian_dic = res[0]
# Get main parameters for ppm scale:
sw = float(varian_dic["procpar"]["sw"]["values"][0]) #direct dimension
sw1 = float(varian_dic["procpar"]["sw1"]["values"][0]) #indirect dimension
obs = float(varian_dic["procpar"]["sfrq"]["values"][0]) #direct dimension
obs1 = float(
varian_dic["procpar"]["dfrq"]["values"][0]) #indirect dimension
car = float(
varian_dic["procpar"]["reffrq"]["values"][0]) #direct dimension
car1 = float(
varian_dic["procpar"]["reffrq1"]["values"][0]) #indirect dimension
# Convert varian to pipe
universal_varian_dic = varian.guess_udic(varian_dic, varian_data)
## direct dimension
universal_varian_dic[0]["sw"] = sw
universal_varian_dic[0]["obs"] = obs
universal_varian_dic[0]["label"] = varian_dic["procpar"]["tn"]["values"][0]
universal_varian_dic[0]["car"] = car
## indirect dimension
universal_varian_dic[1]["sw"] = sw1
universal_varian_dic[1]["obs"] = obs1
universal_varian_dic[1]["label"] = varian_dic["procpar"]["dn"]["values"][0]
universal_varian_dic[1]["car"] = car1
C = convert.converter()
C.from_varian(varian_dic, varian_data, universal_varian_dic)
pipe_dic, varian_pipe_data = C.to_pipe()
## direct dimension
pipe_dic["FDF2SW"] = sw
pipe_dic["FDF2OBS"] = obs
pipe_dic["FDF2LABEL"] = universal_varian_dic[0]["label"]
pipe_dic["FDF2CAR"] = car
## indirect dimension
pipe_dic["FDF1SW"] = sw1
pipe_dic["FDF1OBS"] = obs1
pipe_dic["FDF1LABEL"] = universal_varian_dic[1]["label"]
pipe_dic["FDF1CAR"] = car1
# process the direct dimension
## zero filling
if zero_filling:
pipe_dic, varian_pipe_data = pipe_proc.zf(pipe_dic, varian_pipe_data)
## apodization
if apodization:
lb_d = float(varian_dic["procpar"]["lb"]["values"][0])
pipe_dic, varian_pipe_data = pipe_proc.em(pipe_dic,
varian_pipe_data,
lb=lb_d)
# for lorentz-to-gauss
# pipe_dic, varian_pipe_data = pipe_proc.gm(pipe_dic, varian_pipe_data)
## Fourier transform
pipe_dic, varian_pipe_data = pipe_proc.ft(pipe_dic,
varian_pipe_data,
auto=True)
## Phase Correction
p_zero = float(varian_dic["procpar"]["rp"]["values"][0])
p_one = float(varian_dic["procpar"]["lp"]["values"][0])
pipe_dic, varian_pipe_data = pipe_proc.ps(pipe_dic,
varian_pipe_data,
p0=p_zero,
p1=p_one)
## Remove imaginary numbers:
pipe_dic, varian_pipe_data = pipe_proc.di(pipe_dic, varian_pipe_data)
# process the indirect dimension
pipe_dic, varian_pipe_data = pipe_proc.tp(pipe_dic, varian_pipe_data)
## zero filling
if zero_filling:
pipe_dic, varian_pipe_data = pipe_proc.zf(pipe_dic, varian_pipe_data)
## apodization
if apodization:
lb_d1 = float(varian_dic["procpar"]["lb1"]["values"][0])
pipe_dic, varian_pipe_data = pipe_proc.em(pipe_dic,
varian_pipe_data,
lb=lb_d1)
# for lorentz-to-gauss
# pipe_dic, varian_pipe_data = pipe_proc.gm(pipe_dic, varian_pipe_data)
## Fourier transform
pipe_dic, varian_pipe_data = pipe_proc.ft(pipe_dic,
varian_pipe_data,
auto=True)
## Phase Correction
p_zero1 = float(varian_dic["procpar"]["rp1"]["values"][0])
p_one1 = float(varian_dic["procpar"]["lp1"]["values"][0])
pipe_dic, varian_pipe_data = pipe_proc.ps(pipe_dic,
varian_pipe_data,
p0=p_zero1,
p1=p_one1)
## Remove imaginary numbers:
pipe_dic, varian_pipe_data = pipe_proc.di(pipe_dic, varian_pipe_data)
dic, data = pipe_proc.tp(pipe_dic, varian_pipe_data)
# Calculate both ppm scales:
## direct dimension
ppm = list()
ppm_f = (obs * 1000000 + sw / 2 - (car * 1000000)) / car
ppm_width = sw / car
ppm_i = ppm_f - ppm_width
n = int(data.shape[1])
ppm_step = ppm_width / n
ppm.append(ppm_i)
for i in range(1, n):
ppm.append(ppm[i - 1] + ppm_step)
new_ppm = ppm[::-1]
## indirect dimension
# handling homonuclear cases
if car1 == car and sw == sw1:
ppm1 = list()
ppm1.append(ppm_i)
n1 = int(data.shape[0])
ppm_step1 = ppm_width / n1
for i in range(1, n1):
ppm1.append(ppm1[i - 1] + ppm_step1)
new_ppm1 = ppm1[::-1]
else:
ppm1 = list()
ppm_f1 = (obs1 * 1000000 + sw1 / 2 - (car1 * 1000000)) / car1
ppm_width1 = sw1 / car1
ppm_i1 = ppm_f1 - ppm_width1
n1 = int(data.shape[0])
ppm_step1 = ppm_width1 / n1
ppm1.append(ppm_i1)
for i in range(1, n1):
ppm1.append(ppm1[i - 1] + ppm_step1)
new_ppm1 = ppm1[::-1]
ppms = [new_ppm, new_ppm1]
return (ppms, abs(data))