SPECTRA Module 

klassez.gui.interactive_phase_1D()

getplane(idx, dim='31')[source]

Extract from self.S the idx-th plane (python numbering) on the direction dim. Keeps a reference to it as the attribute self._active.

Parameters:

idxint: Index of the plane to extract
dimstr: '31' = plane F3-F1; '32' = plane F3-F2

Returns:

activepDOSY object: Extracted plane as klassez.Spectra.pDOSY object

integrate(filename=None)[source]

Create the attribute self.D as klassez.fit.DosyFit_pp3D, and integrate the planes along the 31 direction.

Parameters:

filenamestr or None: Custom filename for data and plots. If None, self.filename is used.

Returns:

None

pknl()[source]: Reverses the effect of the digital filter by applying a first order phase correction. To be called after having processed the data by self.process().

plot(dim='31', fqscale=False, Neg=True, lvl0=0.2, name=None)[source]

Plots the real part of the spectrum as a 2D contour plot. You can scroll between the various planes inside the interactive panel.

Parameters:

dimstr: '31' = plane F3-F1; '32' = plane F3-F2
fqscalebool: Display using frequency scale instead of ppm
Negbool: Plot (True) or not (False) the negative contours.
lvl0float: Starting contour value.
namestr: Custom name for the figure. If None, self.filename is used.

process()[source]: Process only the direct dimension. Calls klassez.processing.fp() on each transient. The parameters are read from the procs dictionary.

See also

klassez.processing.fp()

read_procs(other_dir=None)[source]

Reads the procs dictionary from a file named filename.procs in the same directory of the input file.

Parameters:

other_dirstr or None: Different location for the procs dictionary to look into. If None, self.datadir is used instead.

Warning

Do not put the trailing slash!

Returns:

procs: dict: Dictionary of processing parameters

write_procs(other_dir=None)[source]

Writes the actual procs dictionary in a file named "filename.procs" in the same directory of the input file.

Parameters:

other_dir: str or None: Different location for the procs dictionary to write into. If None, self.datadir is used instead.

Warning

Do not put the trailing slash!

class klassez.Spectra.Pseudo_2D(in_file, pv=False, isexp=True)[source]

Bases: Spectrum_2D

Subclass of klassez.Spectra.Spectrum_2D to simulate and handle pseudo-2D experiments. Basically, they share more or less the same attributes, but some methods were adapted in order to suit well with a not-Fourier-transformed indirect dimension.

Attributes:

datadir: str: Path to the input file/dataset directory
filename: str: Base of the name of the file, without extensions
fid: 2darray: FID. For simulated data, this must be explicitely set!
acqus: dict: Dictionary of acqusition parameters
ngdic: dict: Created only if it is an experimental spectrum. Generated by nmrglue.bruker.read(), contains all the information on the spectrometer and on the spectrum.
procs: dict: Dictionary of processing parameters
S: 2darray: Complex spectrum
rr: 2darray: Real part F2, real part F1
ii: 2darray: Imaginary part F2, imaginary part F1
freq_f1: 1darray: Indeces of the experiments, works as placeholder
freq_f2: 1darray: Frequency scale of the direct dimension, in Hz
ppm_f1: 1darray: Indeces of the experiments, works as placeholder
ppm_f2: 1darray: ppm scale of the direct dimension
trf1: dict: Projections of the indirect dimension, as 1darrays . Keys: 'ppm_f2' where they were taken
trf2: dict: Projections of the direct dimension, as 1darrays . Keys: 'ppm_f1' where they were taken
Trf1: dict: Projections of the indirect dimension, as pSpectrum_1D objects. Keys: ‘ppm_f2’ where they were taken
Trf2: dict: Projections of the direct dimension, as pSpectrum_1D objects. Keys: ‘ppm_f1’ where they were taken
integrals: dict: Dictionary where to save the regions and values of the integrals.
F: fit.Voigt_Fit_P2D object: Interface for lineshape deconvolution.

Methods

`abs`([n, lims, alpha, qfil, qfilp])	Correct the baseline automatically on the real part of the spectrum, then retrieve the imaginary part through Hilbert transform.
`abs_v2`([n, lims, alpha, winsize, qfil, qfilp])	Correct the baseline automatically on the real part of the spectrum, then retrieve the imaginary part through Hilbert transform.
`add_noise`([s_n])	Adds noise to the FID, using the function `klassez.sim.noisegen()`.
`adjph`([expno, p0, p1, pv, update])	Adjusts the phases of the spectrum according to the given parameters, or interactively if they are left as default.
`align`([lims, u_off, ref_idx])	Aligns the spectrum to a reference signal in the reference spectrum (default: first one).
`basl`([from_procs, phase])	Apply baseline correction to the whole pseudo-2D by subtracting `self.baseline` from `self.S`.
`cal`([offset, isHz, update, from_procs, ref])	Calibration of the ppm and frequency scales according to a given value, or interactively.
`calf1`([value, isHz, update, from_procs, ref])	Calibrates the ppm and frequency scale of the indirect dimension according to a given value, or interactively.
`calf2`([value, isHz, update, from_procs, ref])	Calibrates the ppm and frequency scale of the direct dimension according to a given value, or interactively.
`convdta`([scaling])	Calls processing.convdta
`eae`()	Calls `klassez.processing.eae()` to shuffle the data and make a States-like FID.
`integrate`([ref, filename, lims, use_bas])	Integrate the spectrum with a dedicated GUI.
`inv_process`()	Performs the inverse processing of the spectrum according to the given parameters.
`mc`()	Computes the magnitude of the spectrum on `self.S`.
`mount`([fids, filename, newacqus])	Replaces the FID of the experiment with a custom one, made by stacking 1D experiments.
`pknl`()	Reverses the effect of the digital filter by applying a first order phase correction.
`plot`([fqscale, Neg, lvl0, Y_label, name])	Plots the real part of the spectrum (`self.rr`).
`plot_md`([which, lims])	Plot a number of experiments, superimposed.
`plot_stacked`([which, lims])	Plot a number of experiments, stacked.
`process`()	Process only the direct dimension.
`projf1`(a[, b])	Calculates the sum trace of the indirect dimension, from `a` ppm to `b` ppm in F2.
`projf2`(a[, b])	Calculates the sum trace of the direct dimension, from `a` ppm to `b` ppm in F1.
`qfil`([which, u, s, from_procs, fqscale])	Gaussian filter to suppress signals.
`read_integrals`(filename[, n])	Reads the integrals in the .igrl file named `filename`, and stores the result in `self.integrals`
`read_procs`([other_dir])	Reads the procs dictionary from a file named "filename.procs" in the same directory of the input file.
`rpbc`([ref_exp])	Computes the phase angles and the baseline using `klassez.processing.rpbc()` on `self.S`.
`scan`([ns, s_n])	Simulates the acquisition of `ns` scans, by adding a different realization of noise at each iteration.
`splitcomb`([J])	Applies the IPAP virtual decoupling scheme by calling klassez.processing.splitcomb`() with appropriate parameters.
`strip`([xlim])	Perform the strip transform in F2 with `xlim`.
`to_wav`([filename, cutoff, rate])	Converts the FID in an audio file by using `klasse.misc.data2wav()`.
`write_acqus`([other_dir])	Write the `acqus` dictionary in a file named "filename.acqus".
`write_integrals`([other_dir])	Write the integrals in a file named `{self.filename}.igrl`.
`write_pdata`([path, procno])	Writes a real/complex array in binary format in `<path>/pdata/<procno>`.
`write_procs`([other_dir])	Writes the actual `procs` dictionary in a file named `"filename.procs"` in the same directory of the input file.
`write_ser`(ser, acqus[, path, filename])	Writes a real/complex array in binary format.
`xf1`()	Process only the indirect dimension.
`xf2`()	Process only the direct dimension.

property S: ndarray[source]: Normal getter for self.S attribute

__init__(in_file, pv=False, isexp=True)[source]

Initialize the class.

Parameters:

in_filestr: path to file to read, or to the folder of the spectrum
pvbool: True if you want to use pseudo-voigt lineshapes for simulation, False for Voigt
isexpbool: True if this is an experimental dataset, False if it is simulated

abs(n=5, lims=None, alpha=2.75, qfil=False, qfilp={'s': 10, 'u': 4.7})[source]

Correct the baseline automatically on the real part of the spectrum, then retrieve the imaginary part through Hilbert transform.

Parameters:

nint

Number of coefficients of the polynomial baseline

limstuple or None

Limits for the region on which to compute the baseline, in ppm

alphafloat

The threshold will be set as thr = alpha * np.std(np.gradient(data))

qfilbool

Choose whether to apply a filter on the solvent region (True) or not (False)

qfilpdict

Parameters to be used to compute the filter if qfil=True. Keys:

‘u’ = center of the filter in ppm
‘s’ = width of the filter in Hz

Returns:

baseline1darray: Employed baseline for correction

See also

klassez.processing.abs()

klassez.processing.abs_v2()

abs_v2(n=5, lims=None, alpha=5, winsize=2, qfil=False, qfilp={'s': 10, 'u': 4.7})[source]

Correct the baseline automatically on the real part of the spectrum, then retrieve the imaginary part through Hilbert transform.

Parameters:

nint: Number of coefficients of the polynomial baseline
limstuple or None: Limits for the region on which to compute the baseline, in ppm
alphafloat: The threshold will be set as thr = alpha * np.std(np.gradient(data))
qfilbool: Choose whether to apply a filter on the solvent region (True) or not (False)
qfilpdict: Parameters to be used to compute the filter if qfil=True. Keys: * ‘u’ = center of the filter in ppm * ‘s’ = width of the filter in Hz

Returns:

baseline1darray: Employed baseline for correction

See also

add_noise(s_n=1)[source]

Adds noise to the FID, using the function klassez.sim.noisegen().

Parameters:

s_nfloat: Standard deviation of the noise

Returns:

None

See also

adjph(expno=0, p0=None, p1=None, pv=None, update=True)[source]

Adjusts the phases of the spectrum according to the given parameters, or interactively if they are left as default.

Parameters:

expno: int: Index of the experiment (python numbering) to use in the interactive panel
p0: float or None: 0-th order phase correction /°
p1: float or None: 1-st order phase correction /°
pv: float or None: 1-st order pivot /ppm
update: bool: Choose if to upload the procs dictionary or not

Returns:

None

See also

klassez.gui.interactive_phase_1D()

align(lims=None, u_off=0.5, ref_idx=0)[source]

Aligns the spectrum to a reference signal in the reference spectrum (default: first one).

Parameters:

lims: tuple or None: Reference signal region, in ppm. If None, you can select it interactively.
u_off: float: Maximum displacement allowed, in ppm
ref_idx: int: Index of the spectrum to be used as a reference (python numbering)

Returns:

None

See also

klassez.gui.get_region()

klassez.processing.align()

basl(from_procs=False, phase=True)[source]

Apply baseline correction to the whole pseudo-2D by subtracting self.baseline from self.S. Then, self.S is unpacked in self.rr and self.ii.

Parameters:

from_procsbool: If True, computes the baseline using the polynomion model reading self.procs['basl_c'] as coefficients
phasebool: Choose if to apply the same phase correction of the spectrum to the baseline. This should be done if the baseline was computed before the phase adjustment!

cal(offset=None, isHz=False, update=True, from_procs=True, *, ref=None)[source]

Calibration of the ppm and frequency scales according to a given value, or interactively. In this latter case, a reference peak must be chosen.

Parameters:

offsetfloat: scale shift F2
isHztuple of bool: True if offset is in frequency units, False if offset is in ppm
updatebool: Choose if to update the procs dictionary or not
from_procsbool: Read the parameters from the procs dictionary
reflist of 1darray: Reference ppm scale and spectrum to be used as reference for calibration [ppm, spectrum]

Returns:

None

See also

klassez.processing.calibrate()

convdta(scaling=1)[source]: Calls processing.convdta

integrate(ref=0, filename=None, lims=None, use_bas=False)[source]

Integrate the spectrum with a dedicated GUI. Calls klassez.anal.integrate_p2D() and writes in self.integrals with keys [ppm1:ppm2] The values are saved in <filename>.igrl.

Parameters:

refint: Index of the transient to be used as reference for the integration.
filenamestr: The integral values will be saved in <filename>.igrl. If None, <self.filename> will be used instead.

Returns:

None

See also

klassez.anal.integrate_p2D()

klassez.processing.stack_fids()

mount(fids=[], filename=None, newacqus=None)[source]

Replaces the FID of the experiment with a custom one, made by stacking 1D experiments. If the default filename exists (i.e. ‘<self.filename>.npy’), the function loads it, otherwise calls klassez.processing.stack_fids() to create it. The fid attribute is overwritten. The key ‘TD1’ of the acqus dictionary is updated to match the first dimension of the new FID.

Parameters:

fidssequence of 1darray or Spectrum_1D objects: FIDs to be stacked. It can be empty if the .npy file already exists.
filenamestr or None: Path to the filename, without the .npy extension. If it is None, the default filename is used.
newacqusdict: New acqus dictionary that replaces the actual one. If it is not a dictionary, no actions are performed.

Returns:

None

See also

pknl()[source]: Reverses the effect of the digital filter by applying a first order phase correction. To be called after having processed the data by self.process().

plot(fqscale=False, Neg=True, lvl0=0.2, Y_label='', name=None)[source]

Plots the real part of the spectrum (self.rr). Use the mouse scroll to adjust the contour starting level.

Parameters:

fqscalebool: Display using frequency scale instead of ppm
Negbool: Plot (True) or not (False) the negative contours.
lvl0float: Starting contour value with respect to the maximum of the spectrum
Y_labelstr: Custom label for the y axis
namestr: Name of the figure. If None, self.filename is used.

plot_md(which=None, lims=None)[source]

Plot a number of experiments, superimposed.

Parameters:

whichstr or None: List of experiment indexes, so that eval(which) is meaningful. None plots all of them
limstuple: Region of the spectrum to show (ppm1, ppm2)

plot_stacked(which=None, lims=None)[source]

Plot a number of experiments, stacked.

Parameters:

whichstr or None: List of experiment indexes, so that eval(which) is meaningful. None plots all of them.
limstuple: Region of the spectrum to show (ppm1, ppm2)

process()[source]: Process only the direct dimension. Calls klassez.processing.fp() on each transient. The parameters are read from the procs dictionary.

See also

klassez.processing.fp()

projf1(a, b=None)[source]

Calculates the sum trace of the indirect dimension, from a ppm to b ppm in F2. Store the trace in the dictionary self.trf1 and as pSpectrum_1D in self.Trf1. The key is 'a' or 'a:b' with two decimal figures.

Calls klassez.anal.get_trace() on self.rr with column=True

Parameters:

afloat: ppm F2 value where to extract the trace.
bfloat or None.: If it is None, extract the trace in a. Else, sum from a to b in F2.

Returns:

None

See also

projf2(a, b=None)[source]

Calculates the sum trace of the direct dimension, from a ppm to b ppm in F1. Store the trace in the dictionary self.trf2 and as pSpectrum_1D in self.Trf2. The key is 'a' or 'a:b' with two decimal figures.

Calls klassez.anal.get_trace() on self.rr with column=False

Parameters:

afloat: ppm F1 value where to extract the trace.
bfloat or None.: If it is None, extract the trace in a. Else, sum from a to b in F1.

Returns:

None

See also

klassez.gui.interactive_qfil()

qfil(which=None, u=None, s=None, from_procs=True, fqscale=False)[source]

Gaussian filter to suppress signals. Tries to read self.procs['qfil'], which is { 'u': u, 's': s } Otherwise, these are set interactively by klassez.gui.interactive_qfil() and then added to self.procs.

Parameters:

whichint or None: Index of the F2 trace to be used for klassez.gui.interactive_qfil(). If None, a suitable trace can be selected using klassez.anal.select_traces().
ufloat: Position /ppm
sfloat: Width (standard deviation) /Hz
from_procsbool: Tries to read the values from self.procs
fqscalebool: True to use the frequency scale, False for the ppm scale.

Returns:

None

See also

klassez.anal.select_traces()

read_integrals(filename, n=-1)[source]

Reads the integrals in the .igrl file named filename, and stores the result in self.integrals

Parameters:

filenamestr: Path to the filename to be read
nint: Number of performed integrating procedure to be read. Default: last one. The breakpoints are lines that start with “!”. For this reason, n=0 returns an empty dictionary, hence the first attempt is n=1.

Returns:

None

See also

klassez.anal.read_igrl()

rpbc(ref_exp=0, **rpbc_kws)[source]

Computes the phase angles and the baseline using klassez.processing.rpbc() on self.S. Then applies the phase correction and subtracts the baseline, automatically, to all experiments of the pseudo-2D. The procs dictionary is then updated and saved. The polynomial baseline is computed according to the given coefficients and stored in self.baseline

Parameters:

ref_exp: int: Index of the reference experiment on which to apply the algorithm
rpbc_kwskeyworded arguments: See klassez.processing.rpbc() for details.

Returns:

None

See also

klassez.processing.rpbc()

scan(ns=1, s_n=1)[source]

Simulates the acquisition of ns scans, by adding a different realization of noise at each iteration. The function is supposed to start with the FID without noise at all. If not, the results will be biased. For reproducible results, fix the seed!

Parameters:

nsint: Number of scans to accumulate
s_nfloat: Standard deviation of the noise for each scan

Returns:

None

See also

klassez.misc.trim_data_2D()

strip(xlim=None)[source]

Perform the strip transform in F2 with xlim. The frequency and ppm scales are updated as well. The pivot for phase correction is moved to the center of the new scale. The imaginary parts are reconstructed via Hilbert transform.

Parameters:

xlimtuple or None: Limits in F2 in ppm

Returns:

None

See also

klassez.processing.hilbert2()

to_wav(filename=None, cutoff=None, rate=44100)[source]

Converts the FID in an audio file by using klasse.misc.data2wav().

Parameters:

filenamestr: Path where to save the file. If None, self.filename is used
cutofffloat: Clipping limits for the FID
rateint: Sampling rate in samples/sec

Returns:

None

See also

klassez.misc.data2wav()

write_integrals(other_dir=None)[source]

Write the integrals in a file named {self.filename}.igrl. The default path is self.datadir.

Parameters:

other_dirstr or None: Different location for the integrals file to write into. If None, self.datadir is used instead.

Returns:

None

See also

klassez.sim.load_sim_1D()

class klassez.Spectra.Spectrum_1D(in_file, pv=False, isexp=True, spect='bruker')[source]

Bases: object

Class: 1D NMR spectrum

Attributes:

datadirstr: Path to the input file/dataset directory
filenamestr: Base of the name of the file, without extensions
fid1darray: FID
acqusdict: Dictionary of acqusition parameters
ngdicdict: Created only if it is an experimental spectrum. Generated by nmrglue, contains all the information on the spectrometer and on the spectrum.
procsdict: Dictionary of processing parameters
S1darray: Normal getter for self.S attribute
r1darray: Real part of the spectrum
i1darray: Imaginary part of the spectrum
freq1darray: Frequency scale of the spectrum, in Hz
ppm1darray: ppm scale of the spectrum
Ffit.Voigt_Fit object: Used for deconvolution. See klassez.fit.Voigt_fit
baseline1darray: Baseline of the spectrum.
integralsdict: Dictionary where to save the regions and values of the integrals.

Methods

`abs`([n, lims, alpha, qfil, qfilp])	Correct the baseline automatically on the real part of the spectrum, then retrieve the imaginary part through Hilbert transform.
`abs_v2`([n, lims, alpha, winsize, qfil, qfilp])	Correct the baseline automatically on the real part of the spectrum, then retrieve the imaginary part through Hilbert transform.
`acme`(**method_kws)	Automatic phase correction based on entropy minimization.
`add_noise`([s_n])	Adds noise to the FID, using the function `klassez.sim.noisegen()`.
`adjph`([p0, p1, pv, update, reference])	Adjusts the phases of the spectrum according to the given parameters, or interactively if they are left as default.
`apk`([alpha, winsize, ap1, seethrough, update])	Automatic phase correction with `klassez.processing.apk()`.
`basl`([from_procs, phase])	Apply the baseline correction by subtracting `self.baseline` from `self.S`.
`blp`([pred, order])	Call `klassez.processing.blp()` on `self.fid` for the application of backward linear prediction to the data.
`cal`([offset, isHz, update, from_procs, ref])	Calibrates the ppm and frequency scale according to a given value, or interactively.
`convdta`([scaling])	Call `klassez.processing.convdta()` using `self.acqus['GRPDLY']`
`integrate`([filename, lims, use_bas])	Integrate the spectrum with a dedicated GUI.
`inv_process`()	Performs the inverse processing of the spectrum according to the given parameters.
`mc`()	Calculates the magnitude of the spectrum and overwrites `self.S`, `self.r`, `self.i`
`pknl`()	Reverses the effect of the digital filter by applying a first order phase correction.
`plot`([fqscale, name])	Plots the real part of the spectrum in an interactive panel for inspection.
`process`([interactive])	Performs the processing of the FID.
`qfil`([u, s, from_procs, fqscale])	Gaussian filter to suppress signals.
`read_integrals`(filename[, n])	Reads the integrals in the .igrl file named `filename`, and stores the result in `self.integrals`
`read_procs`([other_dir])	Reads the procs dictionary from a file named `filename.procs` in the same directory of the input file.
`rpbc`(**rpbc_kws)	Computes the phase angles and the baseline using `klassez.processing.rpbc()` on `self.S`.
`scan`([ns, s_n])	Simulates the acquisition of `ns` scans, by adding a different realization of noise at each iteration.
`to_vf`([filename, Hs, fvf])	Transform a simulated spectrum in a .ivf or .fvf file to be used in a deconvolution procedure.
`to_wav`([filename, cutoff, rate])	Converts the FID in an audio file by using `klasse.misc.data2wav()`.
`write_acqus`([other_dir])	Write the acqus dictionary in a file named `<self.filename>.acqus`.
`write_integrals`([other_dir])	Write the integrals in a file named `{self.filename}.igrl`.
`write_procs`([other_dir])	Writes the actual procs dictionary in a file named `filename.procs` in the same directory of the input file.
`write_ser`(ser, acqus[, path, filename])	Writes a real/complex array in binary format.

Returns:

None

See also

klassez.fit.Voigt_Fit

property S: ndarray[source]: Normal getter for self.S attribute

__init__(in_file, pv=False, isexp=True, spect='bruker')[source]

Initialize the class. Simulation of the dataset (i.e. isexp=False) employs klasez.sim.sim_1D().

Parameters:

in_filestr or dict: path to file to read, or to the folder of the spectrum, or a dictionary of acquisition parameters
pvbool: True if you want to use pseudo-voigt lineshapes for simulation, False for Voigt
isexpbool: True if this is an experimental dataset, False if it is simulated
spectstr: Data file format. Allowed: 'bruker', 'varian', 'magritek', 'oxford', 'jeol'

Returns:

None

See also

klassez.sim.sim_1D()

abs(n=5, lims=None, alpha=2.75, qfil=False, qfilp={'s': 10, 'u': 4.7})[source]

Correct the baseline automatically on the real part of the spectrum, then retrieve the imaginary part through Hilbert transform.

Parameters:

nint

Number of coefficients of the polynomial baseline

limstuple or None

Limits for the region on which to compute the baseline, in ppm

alphafloat

The threshold will be set as thr = alpha * np.std(np.gradient(data))

qfilbool

Choose whether to apply a filter on the solvent region (True) or not (False)

qfilpdict

Parameters to be used to compute the filter if qfil=True. Keys:

‘u’ = center of the filter in ppm
‘s’ = width of the filter in Hz

Returns:

baseline1darray: Employed baseline for correction

See also

klassez.processing.abs()

klassez.processing.abs_v2()

abs_v2(n=5, lims=None, alpha=5, winsize=2, qfil=False, qfilp={'s': 10, 'u': 4.7})[source]

Correct the baseline automatically on the real part of the spectrum, then retrieve the imaginary part through Hilbert transform.

Parameters:

nint: Number of coefficients of the polynomial baseline
limstuple or None: Limits for the region on which to compute the baseline, in ppm
alphafloat: The threshold will be set as thr = alpha * np.std(np.gradient(data))
qfilbool: Choose whether to apply a filter on the solvent region (True) or not (False)
qfilpdict: Parameters to be used to compute the filter if qfil=True. Keys: * ‘u’ = center of the filter in ppm * ‘s’ = width of the filter in Hz

Returns:

baseline1darray: Employed baseline for correction

See also

klassez.processing.acme()

acme(**method_kws)[source]

Automatic phase correction based on entropy minimization. It calculates the phase angles using klassez.processing.acme() specified in method, then calls self.adjph() with those values.

Parameters:

method_kwskeyworded arguments: Additional parameters for the chosen method.

Returns:

None

See also

add_noise(s_n=1)[source]

Adds noise to the FID, using the function klassez.sim.noisegen().

Parameters:

s_nfloat: Standard deviation of the noise

Returns:

None

See also

adjph(p0=None, p1=None, pv=None, update=True, reference=None)[source]

Adjusts the phases of the spectrum according to the given parameters, or interactively if they are left as default. Calls for klassez.processing.ps().

Parameters:

p0float or None: 0-th order phase correction /°
p1float or None: 1-st order phase correction /°
pvfloat or None: 1-st order pivot /ppm
updatebool: Choose if you want to update the procs dictionary or not
referencelist of 1darray or Spectrum_1D object: Reference spectrum to be used for phasing. Can be also given as [ppm, spectrum]

Returns:

valuestuple: (p0, p1, pivot)

See also

klassez.gui.interactive_phase_1D()

apk(alpha=3, winsize=50, ap1=True, seethrough=False, update=True)[source]

Automatic phase correction with klassez.processing.apk().

Parameters:

alphafloat: Factor that multiplies the std of the spectrum to set the threshold
winsizefloat: Minimum size of the window that can contain peaks /Hz
ap1bool: True to adjust both zero and first order, False for only phase zero
seethroughbool: If True, draws a series of diagnostic figures to see what the algorithm is doing
updatebool: Update the procs dictionary

Returns:

None

See also

klassez.processing.apk()

basl(from_procs=False, phase=True)[source]

Apply the baseline correction by subtracting self.baseline from self.S. Then, self.S is unpacked in self.r and self.i

Parameters:

from_procsbool: If True, computes the baseline using the polynomion model reading self.procs['basl_c'] as coefficients
phasebool: Choose if to apply the same phase correction of the spectrum to the baseline. This should be done if the baseline was computed before the phase adjustment!

blp(pred=8, order=8)[source]

Call klassez.processing.blp() on self.fid for the application of backward linear prediction to the data. Important for Oxford benchtop data, where you have to predict 8 points to have a usable spectrum.

Parameters:

predint: Number of points to be predicted
orderint: Number of coefficients to be used for the prediction
Nint: Number of FID points to be used for calculation; used to decrease computation time

Returns:

None

See also

klassez.processing.blp()

cal(offset=None, isHz=False, update=True, from_procs=True, *, ref=None)[source]

Calibrates the ppm and frequency scale according to a given value, or interactively. Calls klassez.processing.calibration().

Parameters:

offsetfloat or None: scale shift value
isHzbool: True if offset is in frequency units, False if offset is in ppm
updatebool: Choose if to update the procs dictionary or not
from_procsbool: Read the parameters from the procs dictionary
reflist of 1darray or Spectrum_1D object: Reference spectrum to be used for calibration. If list, [ppm scale, spectrum]

Returns:

None

See also

klassez.processing.calibration()

convdta(scaling=1)[source]: Call klassez.processing.convdta() using self.acqus['GRPDLY']

Note

It is better to use klassez.Spectra.Spectrum_1D.pknl(), convdta often causes baseline distortions

See also

klassez.processing.convdta()

integrate(filename=None, lims=None, use_bas=False)[source]

Integrate the spectrum with a dedicated GUI. Calls klassez.anal.integrate() and writes in self.integrals with keys [ppm1:ppm2] The values are saved in <filename>.igrl.

Parameters:

filenamestr: The integral values will be saved in <filename>.igrl. If None, <self.filename> will be used instead.

Returns:

None

See also

klassez.anal.integrate()

klassez.processing.fp()

inv_process()[source]: Performs the inverse processing of the spectrum according to the given parameters. Overwrites the S attribute!! Calls klassez.processing.inv_fp().

See also

klassez.processing.inv_fp()

mc()[source]: Calculates the magnitude of the spectrum and overwrites self.S, self.r, self.i

\[\text{mag}_S = \sqrt{\Re\{S\}^2 + \Im\{S\}^2}\]

pknl()[source]: Reverses the effect of the digital filter by applying a first order phase correction. If the S attribute exists, the processing is applied on the processed data. Otherwise, the correction is performed on the FID. This function calls klassez.processing.pknl() with the parameter grpdly read from the acqus dictionary.

See also

klassez.processing.pknl()

plot(fqscale=False, name=None)[source]

Plots the real part of the spectrum in an interactive panel for inspection.

Parameters:

fqscalebool: Display using frequency scale instead of ppm
namestr: Filename for the figure. If None, self.filename is used

process(interactive=False)[source]

Performs the processing of the FID. The parameters are read from self.procs. Calls klassez.gui.interactive_fp() or klassez.processing.fp() using self.acqus and self.procs Writes the result is self.S, then unpacks it in self.r and self.i Calculates frequency and ppm scales. Also initializes self.F with klassez.fit.Voigt_Fit class using the current parameters.

Parameters:

interactivebool: True if you want to open the interactive panel, False to read the parameters from self.procs.

Returns:

None

See also

klassez.gui.interactive_fp()

klassez.fit.Voigt_Fit

qfil(u=None, s=None, from_procs=True, fqscale=False)[source]

Gaussian filter to suppress signals. Tries to read self.procs['qfil'], which is

{ 'u': u, 's': s }

Otherwise, these are set interactively by klassez.gui.interactive_qfil() and then added to self.procs.

Parameters:

ufloat: Position of the filter /ppm
sfloat: Width (standard deviation) of the filter /ppm
from_procsbool: Tries to read the values from self.procs
fqscalebool: True to use the frequency scale, False for the ppm scale.

Returns:

None

See also

klassez.gui.interactive_qfil()

read_integrals(filename, n=-1)[source]

Reads the integrals in the .igrl file named filename, and stores the result in self.integrals

Parameters:

filenamestr: Path to the filename to be read
nint: Number of performed integrating procedure to be read. Default: last one. The breakpoints are lines that start with “!”. For this reason, n=0 returns an empty dictionary, hence the first attempt is n=1.

Returns:

None

See also

klassez.anal.read_igrl()

read_procs(other_dir=None)[source]

Reads the procs dictionary from a file named filename.procs in the same directory of the input file.

Parameters:

other_dirstr or None: Different location for the procs dictionary to look into. If None, self.datadir is used instead.

Warning

Do not put the trailing slash!

Returns:

procs: dict: Dictionary of processing parameters

rpbc(**rpbc_kws)[source]

Computes the phase angles and the baseline using klassez.processing.rpbc() on self.S. Then applies the phase correction and subtracts the baseline, automatically. The procs dictionary is then updated and saved. The polynomial baseline is computed according to the given coefficients and stored in self.baseline

Parameters:

rpbc_kwskeyworded arguments: See klassez.processing.rpbc() for details.

Returns:

None

See also

klassez.processing.rpbc()

scan(ns=1, s_n=1)[source]

Simulates the acquisition of ns scans, by adding a different realization of noise at each iteration. The function is supposed to start with the FID without noise at all. If not, the results will be biased. For reproducible results, fix the seed!

Parameters:

nsint: Number of scans to accumulate
s_nfloat: Standard deviation of the noise for each scan

Returns:

None

See also

klassez.misc.data2wav()

to_vf(filename=None, Hs=None, fvf=True)[source]

Transform a simulated spectrum in a .ivf or .fvf file to be used in a deconvolution procedure. To do this, it reads the peak parameters saved in self.acqus. The number of signals is determined by acqus['amplitudes']. Multiplets are splitted according to their Js, and saved as components.

Parameters:

filenamestr or None: Path to the filename to be saved, without extension. If None, self.filename is used
Hsint or None: Number of nuclei the spectrum integrates for. If None, the sum of the amplitudes is used.
fvfbool: If True, adds the ‘.fvf’ extension to the filename, if False, adds ‘.ivf’

to_wav(filename=None, cutoff=None, rate=44100)[source]

Converts the FID in an audio file by using klasse.misc.data2wav().

Parameters:

filenamestr: Path where to save the file. If None, self.filename is used
cutofffloat: Clipping limits for the FID
rateint: Sampling rate in samples/sec

Returns:

None

See also

write_acqus(other_dir=None)[source]

Write the acqus dictionary in a file named <self.filename>.acqus. Calls klassez.misc.write_acqus_1D()

Parameters:

other_dirstr or None: Different location for the acqus dictionary to write into. If None, self.datadir is used instead.

Returns:

None

See also

klassez.misc.write_acqus_1D()

write_integrals(other_dir=None)[source]

Write the integrals in a file named {self.filename}.igrl. The default path is self.datadir.

Parameters:

other_dirstr or None: Different location for the integrals file to write into. If None, self.datadir is used instead.

Returns:

None

See also

klassez.misc.write_ser()

write_procs(other_dir=None)[source]

Writes the actual procs dictionary in a file named filename.procs in the same directory of the input file.

Parameters:

other_dirstr or None: Different location for the procs dictionary to write into. If None, self.datadir is used instead.

Warning

Do not put the trailing slash!

static write_ser(ser, acqus, path=None, filename=None)[source]

Writes a real/complex array in binary format. Calls klassez.misc.write_ser`(). Be sure that acqus contains the BYTORDA and DTYPA keys.

Parameters:

serndarray: Array that you want to convert in binary format.
acqusdict: Dictionary of acquisition parameters. It must contain BYTORDA and DTYPA.
pathstr: Path where to save the binary file.

Returns:

None

See also

class klassez.Spectra.Spectrum_2D(in_file, spect='bruker', pv=False, isexp=True)[source]

Bases: object

Class: 2D NMR spectrum

Attributes:

datadirstr: Path to the input file/dataset directory
filenamestr: Base of the name of the file, without extensions
fid2darray: FID
acqusdict: Dictionary of acqusition parameters
ngdicdict: Created only if it is an experimental spectrum. Generated by nmrglue.<spectrometer>.read(), contains all the information on the spectrometer and on the spectrum.
procsdict: Dictionary of processing parameters
eaeflagint: If FnMODE='Echo-Antiecho', keeps track of the manipulation of the data so to not repeat the same process twice
S2darray: Normal getter for self.S attribute
rr2darray: Real part F2, real part F1
ii2darray: Imaginary part F2, imaginary part F1
ir2darray: Real part F2, imaginary part F1. Only exist if F1 is acquired in phase-sensitive mode
ri2darray: Imaginary part F2, real part F1. Only exist if F1 is acquired in phase-sensitive mode
freq_f11darray: Frequency scale of the indirect dimension, in Hz
freq_f21darray: Frequency scale of the direct dimension, in Hz
ppm_f11darray: ppm scale of the indirect dimension
ppm_f21darray: ppm scale of the direct dimension
trf1dict: Projections of the indirect dimension, as 1darrays. Keys: ‘ppm_f2’ where they were taken
trf2dict: Projections of the direct dimension, as 1darrays. Keys: ‘ppm_f1’ where they were taken
Trf1dict: Projections of the indirect dimension, as pSpectrum_1D objects. Keys: ‘ppm_f2’ where they were taken
Trf2dict: Projections of the direct dimension, as pSpectrum_1D objects. Keys: ‘ppm_f1’ where they were taken
integralsdict: Dictionary where to save the regions and values of the integrals.

Methods

`add_noise`([s_n])	Adds noise to the FID, using the function `klassez.sim.noisegen()`.
`adjph`([p01, p11, pv1, p02, p12, pv2, update])	Adjusts the phases of the spectrum according to the given parameters, or interactively if they are left as default.
`cal`([offset, isHz, update, from_procs, ...])	Calibration of the ppm and frequency scales according to a given value, or interactively.
`calf1`([value, isHz, update, from_procs, ref])	Calibrates the ppm and frequency scale of the indirect dimension according to a given value, or interactively.
`calf2`([value, isHz, update, from_procs, ref])	Calibrates the ppm and frequency scale of the direct dimension according to a given value, or interactively.
`convdta`([scaling])	Calls `klassez.processing.convdta()` to compensate for the group delay.
`eae`()	Calls `klassez.processing.eae()` to shuffle the data and make a States-like FID.
`integrate`(**kwargs)	Integrates the spectrum with a dedicated GUI.
`inv_process`()	Performs the inverse processing of the spectrum according to the given parameters.
`mc`()	Computes the magnitude of the spectrum on `self.S`.
`pknl`()	Reverses the effect of the digital filter by applying a first order phase correction.
`plot`([fqscale, Neg, lvl0, name])	Plots the real part of the spectrum (`self.rr`).
`process`([interactive])	Performs the full processing of the FID on both dimensions.
`projf1`(a[, b])	Calculates the sum trace of the indirect dimension, from `a` ppm to `b` ppm in F2.
`projf2`(a[, b])	Calculates the sum trace of the direct dimension, from `a` ppm to `b` ppm in F1.
`qfil`([which, u, s, from_procs, fqscale, ...])	Gaussian filter to suppress signals.
`read_procs`([other_dir])	Reads the procs dictionary from a file named "filename.procs" in the same directory of the input file.
`scan`([ns, s_n])	Simulates the acquisition of `ns` scans, by adding a different realization of noise at each iteration.
`splitcomb`([J])	Applies the IPAP virtual decoupling scheme by calling klassez.processing.splitcomb`() with appropriate parameters.
`strip`([xlim, ylim])	Perform the strip transform in F2 with `xlim` and on F1 with `ylim`.
`to_wav`([filename, cutoff, rate])	Converts the FID in an audio file by using `klassez.misc.data2wav()`.
`write_acqus`([other_dir])	Write the `acqus` dictionary in a file named "filename.acqus".
`write_integrals`([other_dir])	Write the integrals in a file named `"{self.filename}.int"`.
`write_pdata`([path, procno])	Writes a real/complex array in binary format in `<path>/pdata/<procno>`.
`write_procs`([other_dir])	Writes the actual `procs` dictionary in a file named `"filename.procs"` in the same directory of the input file.
`write_ser`(ser, acqus[, path, filename])	Writes a real/complex array in binary format.
`xf1`()	Process only the indirect dimension.
`xf2`()	Process only the direct dimension.

property S: ndarray[source]: Normal getter for self.S attribute

__init__(in_file, spect='bruker', pv=False, isexp=True)[source]

Initialize the class.

Parameters:

in_filestr: path to file to read, or to the folder of the spectrum
spectstr: Data file format. Allowed: 'bruker' only
pvbool: True if you want to use pseudo-voigt lineshapes for simulation, False for Voigt
isexpbool: True if this is an experimental dataset, False if it is simulated

add_noise(s_n=1)[source]

Adds noise to the FID, using the function klassez.sim.noisegen().

Parameters:

s_nfloat: Standard deviation of the noise

adjph(p01=None, p11=None, pv1=None, p02=None, p12=None, pv2=None, update=True)[source]

Adjusts the phases of the spectrum according to the given parameters, or interactively if they are left as default.

The non-interactive workflow is to apply klassez.processing.ps() on F2, transpose according to FnMODE, apply klassez.processing.ps() on F1, transpose back. If FnMODE='No', the phase correction is applied only on F2, as it should be done in a pseudo-2D experiment. Once self.S was updated and unpacked, the phase values are added to the procs dictionary to keep track of multiple phase adjustments.

Parameters:

p01float or None: 0-th order phase correction /° of the indirect dimension
p11float or None: 1-st order phase correction /° of the indirect dimension
pv1float or None: 1-st order pivot /ppm of the indirect dimension
p02float or None: 0-th order phase correction /° of the direct dimension
p12float or None: 1-st order phase correction /° of the direct dimension
pv2float or None: 1-st order pivot /ppm of the direct dimension
updatebool: Choose if to update the procs dictionary or not

Returns:

None

See also

klassez.gui.interactive_phase_2D()

cal(offset=[None, None], isHz=False, update=True, from_procs=True, *, ref1=None, ref2=None)[source]

Calibration of the ppm and frequency scales according to a given value, or interactively. In this latter case, a reference peak must be chosen.

Parameters:

offsettuple: (scale shift F1, scale shift F2)
isHztuple of bool: True if offset is in frequency units, False if offset is in ppm
updatebool: Choose if to update the procs dictionary or not
from_procsbool: Read the parameters from the procs dictionary
ref1list of 1darray: Reference ppm scale and spectrum to be used as reference for calibration of F1 [ppm, spectrum]
ref2list of 1darray: Reference ppm scale and spectrum to be used as reference for calibration of F2 [ppm, spectrum]

Returns:

None

See also

klassez.processing.calibration()

klassez.anal.select_traces()

calf1(value=None, isHz=False, update=True, from_procs=True, *, ref=None)[source]

Calibrates the ppm and frequency scale of the indirect dimension according to a given value, or interactively. Calls self.cal() on F1 only.

Parameters:

valuefloat or None: scale shift value
isHzbool: True if offset is in frequency units, False if offset is in ppm
updatebool: Choose if to update the procs dictionary or not
from_procsbool: Read the parameters from the procs dictionary
reflist of 1darray: Reference ppm scale and spectrum to be used as reference for calibration [ppm, spectrum]

Returns:

None

See also

klassez.Spectra.Spectrum_2D.cal()

calf2(value=None, isHz=False, update=True, from_procs=True, *, ref=None)[source]

Calibrates the ppm and frequency scale of the direct dimension according to a given value, or interactively. Calls self.cal() on F2 only

Parameters:

valuefloat or None: scale shift value
isHzbool: True if offset is in frequency units, False if offset is in ppm
updatebool: Choose if to update the procs dictionary or not
from_procsbool: Read the parameters from the procs dictionary
reflist of 1darray: Reference ppm scale and spectrum to be used as reference for calibration [ppm, spectrum]

Returns:

None

See also

klassez.Spectra.Spectrum_2D.cal()

convdta(scaling=1)[source]

Calls klassez.processing.convdta() to compensate for the group delay. It does not always work, depends on TopSpin version and planets alignment.

Warning

Better to use klassez.Spectra.Spectrum_2D.pknl()

Parameters:

scalingfloat: Scaling factor for processingconvdta.

eae()[source]: Calls klassez.processing.eae() to shuffle the data and make a States-like FID. Sets self.eaeflag to 0.

integrate(**kwargs)[source]

Integrates the spectrum with a dedicated GUI. Stores the results in self.integral. Calls klassez.anal.integrate_2D()

Parameters:

kwargs: keyworded arguments: Additional parameters for klassez.anal.integrate_2D()

Returns:

None

See also

klassez.anal.integrate_2D()

inv_process()[source]: Performs the inverse processing of the spectrum according to the given parameters. Overwrites the S attribute!! Calls klassez.processing.inv_xfb().

mc()[source]: Computes the magnitude of the spectrum on self.S. Then, updates rr, ri, ir, ii.

See also

klassez.processing.unpack_2D()

pknl()[source]: Reverses the effect of the digital filter by applying a first order phase correction. If the S attribute exists, the processing is applied on the processed data. Otherwise, the correction is performed on the FID. This function calls klassez.processing.pknl() with the parameter grpdly read from the acqus dictionary.

See also

klassez.processing.pknl()

plot(fqscale=False, Neg=True, lvl0=0.2, name=None)[source]

Plots the real part of the spectrum (self.rr). Use the mouse scroll to adjust the contour starting level.

Parameters:

fqscalebool: Display using frequency scale instead of ppm
Negbool: Plot (True) or not (False) the negative contours.
lvl0float: Starting contour value with respect to the maximum of the spectrum
namestr: Name of the figure. If None, self.filename is used.

process(interactive=False, **int_kwargs)[source]

Performs the full processing of the FID on both dimensions. The parameters are read from self.procs. If FnMODE='Echo-Antiecho' and you did not call self.eae() before, the FID is converted to States-TPPI with klassez.processing.eae() before to start. If interactive=True, calls klassez.gui.interactive_xfb() with int_kwargs, else calls klassez.processing.xfb().

The complex/hypercomplex spectrum is stored in self.S, then unpacked into self.rr, self.ri, self.ir, self.ii. If FnMODE='Echo-Antiecho', a phase correction of -90 degrees is applied on the indirect dimension.

Parameters:

interactivebool: True if you want to open the interactive panel, False to read the parameters from self.procs.
int_kwargskeyworded arguments: Additional parameters for klassez.gui.interactive_xfb(), if interactive=True.

Returns:

None

See also

klassez.processing.eae()

klassez.gui.interactive_xfb()

klassez.processing.xfb()

klassez.processing.unpack_2D()

projf1(a, b=None)[source]

Calculates the sum trace of the indirect dimension, from a ppm to b ppm in F2. Store the trace in the dictionary self.trf1 and as pSpectrum_1D in self.Trf1. The key is 'a' or 'a:b' with two decimal figures.

Calls klassez.anal.get_trace() on self.rr with column=True

Parameters:

afloat: ppm F2 value where to extract the trace.
bfloat or None.: If it is None, extract the trace in a. Else, sum from a to b in F2.

Returns:

None

See also

projf2(a, b=None)[source]

Calculates the sum trace of the direct dimension, from a ppm to b ppm in F1. Store the trace in the dictionary self.trf2 and as pSpectrum_1D in self.Trf2. The key is 'a' or 'a:b' with two decimal figures.

Calls klassez.anal.get_trace() on self.rr with column=False

Parameters:

afloat: ppm F1 value where to extract the trace.
bfloat or None.: If it is None, extract the trace in a. Else, sum from a to b in F1.

Returns:

None

See also

klassez.gui.interactive_qfil()

qfil(which=None, u=None, s=None, from_procs=True, fqscale=False, *, SFO=None, O1P=None)[source]

Gaussian filter to suppress signals. Tries to read self.procs['qfil'], which is { 'u': u, 's': s } Otherwise, these are set interactively by klassez.gui.interactive_qfil() and then added to self.procs.

Parameters:

whichint or None: Index of the F2 trace to be used for klassez.gui.interactive_qfil(). If None, a suitable trace can be selected using klassez.anal.select_traces().
ufloat: Position /ppm
sfloat: Width (standard deviation) /Hz
from_procsbool: Read the parameters from the procs dictionary
fqscalebool: True to use the frequency scale, False for the ppm scale.
SFOfloat: Nucleus Larmor frequency to use for the conversion to Hz. If None, self.acqus['SFO2'] is used by default.
O1Pfloat: Carrier position in ppm. If None, self.acqus['o2p'] is used by default.

Returns:

None

See also

klassez.anal.select_traces()

read_procs(other_dir=None)[source]

Reads the procs dictionary from a file named “filename.procs” in the same directory of the input file.

Parameters:

other_dirstr or None: Different location for the procs dictionary to look into. If None, self.datadir is used instead.

Warning

Do not put the trailing slash!

Returns:

procsdict: Dictionary of processing parameters

scan(ns=1, s_n=1)[source]

Simulates the acquisition of ns scans, by adding a different realization of noise at each iteration. The function is supposed to start with the FID without noise at all. If not, the results will be biased. For reproducible results, fix the seed!

Parameters:

nsint: Number of scans to accumulate
s_nfloat: Standard deviation of the noise for each scan

Returns:

None

See also

klassez.processing.splitcomb()

splitcomb(J=53.8)[source]

Applies the IPAP virtual decoupling scheme by calling klassez.processing.splitcomb`() with appropriate parameters. Replaces the FID.

Parameters:

Jfloat: Scalar coupling constant, in Hz, of the coupling to be suppressed.

Returns:

None

See also

strip(xlim=None, ylim=None)[source]

Perform the strip transform in F2 with xlim and on F1 with ylim. The frequency and ppm scales are updated as well. The pivot for phase correction is moved to the center of the new scale. The imaginary parts are reconstructed via Hilbert transform.

Parameters:

xlimtuple or None: Limits in F2 in ppm
ylimtuple or None: Limits in F1 in ppm

Returns:

None

See also

klassez.misc.trim_data_2D()