ANAL Module

This package contains function for the analysis of spectra.

klassez.anal.get_trace(data, ppm_f2, ppm_f1, a, b=None, column=True)

Takes as input a 2D dataset and the ppm scales of direct and indirect dimensions respectively. Calculates the projection on the given axis summing from a (ppm) to b (ppm). Default: indirect dimension projection (i.e. column=True), change it to False for the direct dimension projection.

Parameters:

data2darray

Spectrum of which to extract the projections

ppm_f21darray

ppm scale of the direct dimension

ppm_f11darray

ppm scale of the indirect dimension

afloat

The ppm value from which to start extracting the projection.

bfloat, optional

If provided, the ppm value at which to stop extracting the projection. Otherwise, returns only the a trace.

columnbool

If True, extracts the F1 projection. If False, extracts the F2 projection.

Returns:

y1darray

Computed projection

klassez.anal.integral_2D(ppm_f1, t_f1, SFO1, ppm_f2, t_f2, SFO2, u_1=None, fwhm_1=200, utol_1=0.5, u_2=None, fwhm_2=200, utol_2=0.5, plot_result=False)

Calculate the integral of a 2D peak. The idea is to extract the traces correspondent to the peak center and fit them with a gaussian function in each dimension. Then, once got the intensity of each of the two gaussians, multiply them together in order to obtain the 2D integral. This procedure should be equivalent to what CARA does.

Note

In development!!!

Parameters:

ppm_f11darray

PPM scale of the indirect dimension

t_f11darray

Trace of the indirect dimension, real part

SFO1float

Larmor frequency of the nucleus in the indirect dimension

ppm_f21darray

PPM scale of the direct dimension

t_f21darray

Trace of the direct dimension, real part

SFO2float

Larmor frequency of the nucleus in the direct dimension

u_1float

Chemical shift in F1 /ppm. Defaults to the center of the scale

fwhm_1float

Starting FWHM /Hz in the indirect dimension

utol_1float

Allowed tolerance for u_1 during the fit. (u_1-utol_1, u_1+utol_1)

u_2float

Chemical shift in F2 /ppm. Defaults to the center of the scale

fwhm_2float

Starting FWHM /Hz in the direct dimension

utol_2float

Allowed tolerance for u_2 during the fit. (u_2-utol_2, u_2+utol_2)

plot_resultbool

True to show how the program fitted the traces.

Returns:

I_totfloat

Computed integral.

klassez.anal.integrate(ppm0, data0, SFO1, filename='integrals', X_label='$\\delta\\,$F1 /ppm', dx=1)

Allows interactive integration of a NMR spectrum through a dedicated GUI. Returns the values as a dictionary, where the keys are the selected regions truncated to the 3nd decimal figure. The values are saved in the <filename>.igrl file.

In the GUI, draw the integration region around the peak. The shape of the integral function appears in red in that region, and the value of the integral is reported under the “current integral” label on the right side. The integral can be corrected with a baseline, that is the straight line that connects the border of the integration window. It can be activated and deactivated using the checkbox on the right side. Press “ADD” to save the integral value. Upon pressing, the integral function becomes green and the value appears on top of the figure. To remove an integral from the list, first click on the corresponding value on the top to select it: the number and the integral function should become purple. Then, click on the “REMOVE” button. Use the right click to reset the selection.

At the end of the procedure, click on the “SAVE” button to write the .igrl file.

Parameters:

ppm1darray

PPM scale of the spectrum

data1darray

Spectrum to be integrated.

SFO1float

Larmor frequency of the observed nucleus. For conversion to the correct integration scale.

filenamestr

Name for the .igrl file to be written. Without extension!

X_labelstr

Label of the x-axis

dxfloat

Correction for the integral values. It should be dx = 2 * dw

Returns:

abs_valsdict

Dictionary containing the values of the integrated peaks.

See also

klassez.anal.write_igrl()

klassez.anal.integrate_2D(ppm_f1, ppm_f2, data, SFO1, SFO2, fwhm_1=200, fwhm_2=200, utol_1=0.5, utol_2=0.5, plot_result=False)

Function to select and integrate 2D peaks of a spectrum, using dedicated GUIs. Calls integral_2D to do the dirty job.

Error

Old function!! Legacy

Parameters:

ppm_f11darray

PPM scale of the indirect dimension

ppm_f21darray

PPM scale of the direct dimension

data2darray

real part of the spectrum

SFO1float

Larmor frequency of the nucleus in the indirect dimension

SFO2float

Larmor frequency of the nucleus in the direct dimension

fwhm_1float

Starting FWHM /Hz in the indirect dimension

fwhm_2float

Starting FWHM /Hz in the direct dimension

utol_1float

Allowed tolerance for u_1 during the fit. (u_1-utol_1, u_1+utol_1)

utol_2float

Allowed tolerance for u_2 during the fit. (u_2-utol_2, u_2+utol_2)

plot_resultbool

True to show how the program fitted the traces.

Returns:

Idict

Computed integrals. The keys are '<ppm f1>:<ppm f2>' with 2 decimal figures.

klassez.anal.integrate_p2D(ppm0, data0, SFO1, ref=0, indirect_scale=None, filename='integrals', X_label='$\\delta\\,$F1 /ppm', dx=1)

Allows interactive integration of a (series of) NMR spectrum through a dedicated GUI. Returns the values as a dictionary, where the keys are the selected regions truncated to the 3nd decimal figure. The values are saved in the <filename>.igrl file.

In the GUI, draw the integration region around the peak. The shape of the integral function appears in red in that region, and the value of the integral is reported under the “current integral” label on the right side. Both are referred to the ref-th row of data. The trend of the integrals appear in the inset plot at the bottom left corner. The integral can be corrected with a baseline, that is the straight line that connects the border of the integration window. It can be activated and deactivated using the checkbox on the right side. Press “ADD” to save the integral values. Upon pressing, the integral function becomes green and the value appears on top of the figure. To remove an integral from the list, first click on the corresponding value on the top to select it: the number and the integral function should become purple. Then, click on the “REMOVE” button. Use the right click to reset the selection.

At the end of the procedure, click on the “SAVE” button to write the .igrl file.

Parameters:

ppm1darray

PPM scale of the spectrum

data1darray

Spectrum to be integrated.

SFO1float

Larmor frequency of the observed nucleus. For conversion to the correct integration scale.

refint

Index of the transient to be used as reference when drawing the functions

indirect_scale1darray

Scale of the indirect dimension, if any. If None, 1, 2, 3 … are used.

filenamestr

Name for the .igrl file to be written. Without extension!

X_labelstr

Label of the x-axis

dxfloat

Correction for the integral values. It should be dx = 2 * dw

Returns:

abs_valsdict

Dictionary containing the values of the integrated peaks.

See also

klassez.anal.write_igrl()

klassez.anal.noise_std(y)

Calculates the standard deviation of the noise using the Bruker formula:

Taken \(y\) as an array of \(N\) points, and \(y[i]\) its i-th entry:

\[\sigma_N = \frac{1}{\sqrt{r-1}} \sqrt{ \sum_{k=0}^{r-1} (y[k]^2) - \frac{1}{r} [ ( \sum_{k=0}^{r-1} y[k] )^2 + \frac{3}{r^2 -1}( \sum_{k=0}^{r / 2 - 1} (k+1) (y[r/ 2 + k] - y[r/ 2 - k -1 ] ) )^2 ] }\]

Parameters:

y1darray

The spectral region you would like to use to calculate the standard deviation of the noise.

Returns:

noisestdfloat

The standard deviation of the noise.

klassez.anal.read_igrl(filename, n=-1)

Reads a .igrl file, containing the integrals of either one or a series of spectra. The file is separated and unpacked into a dictionary, each of which contains the integration windows with three decimal figures as keys and the integrals as their associated value. If present, the indirect timescale is also read and returned.

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:

dicdict

Integrals

indirect_scale1darray or None

If there is a line that starts with 'x = ', it is interpreted as the scale for the indirect dimension. If there is not, None is returned.

See also

klassez.fit.write_igrl()

klassez.anal.select_for_integration(ppm_f1, ppm_f2, data, Neg=True)

Select the peaks of a 2D spectrum to integrate. First, select the area where your peak is located by dragging the red square. Then, select the center of the peak by right_clicking. Finally, click ‘ADD’ to store the peak. Repeat the procedure for as many peaks as you want.

Parameters:

ppm_f11darray

ppm scale of the indirect dimension

ppm_f21darray

ppm scale of the direct dimension

data2darray

Spectrum

Negbool

Choose if to show the negative contours ( True) or not ( False )

Returns:

peakslist of dict

For each peak there are two keys, ‘f1’ and ‘f2’, whose meaning is obvious. For each of these keys, you have ‘u’: center of the peak /ppm, and ‘lim’: the limits of the square you drew before.

klassez.anal.select_traces(ppm_f1, ppm_f2, data, Neg=True)

Select traces from a 2D spectrum, save the coordinates in a list. Left click to select a point, right click to remove it.

Parameters:

ppm_f11darray

ppm scale of the indirect dimension

ppm_f21darray

ppm scale of the direct dimension

data2darray

Spectrum

Negbool

Choose if to show the negative contours ( True) or not ( False )

Returns:

coordlist

List containing the [x,y] coordinates of the selected points.

klassez.anal.snr(x, data, s_reg=None, n_reg=None, gui=False)

Computes the signal to noise ratio of a 1D spectrum as height of the signal over twice the noise standard deviation. If you are not sure about what to put into s_reg and n_reg, use the interactive interface by setting gui=True. The values will be suggested.

Parameters:

x1darray

Scale of the spectrum to use. The values in n_reg are searched according to this scale

data1darray

The spectrum of which you want to compute the SNR

s_regfloat or tuple

If float, uses this value as maximum signal. If tuple, the maximum value within the s_reg region is used as maximum signal. In this case, the limits of such region must be given according to x. If None, it is selected as the maximum value in data

n_reglist or tuple, optional

If provided, contains the points that delimit the noise region. Otherwise, the whole spectrum is used.

guibool

Set it to True to ignore s_reg and n_reg, and use the GUI for the interactive estimation.

Returns:

snrfloat

The SNR of the spectrum

See also

klassez.anal.noise_std() klassez.anal.snr_gui()

klassez.anal.snr_2D(x_f1, x_f2, data, s_reg=None, n_reg=None, gui=False)

Computes the signal to noise ratio of a 2D spectrum as the height of the signal divided by twice the standard deviation of the noise, for each dimension. If you are not sure about what to put into s_reg and n_reg, use the interactive interface by setting gui=True. The values will be suggested.

Parameters:

x_f11darray

PPM scale of the spectrum for the indirect dimension

x_f21darray

PPM scale of the spectrum for the direct dimension

data2darray

Spectrum of which you want to compute the SNR

s_regfloat or list of tuple

If float, uses this value as maximum signal. If None, it is selected as the maximum value in data Otherwise, you have to provide the limits to draw a rectangle on data as [(f2_sx, f2_dx), (f1_sx, f1_dx)]; the maximum signal is the highest point within this region.

n_reglist or tuple

Contains the coordinates at which to extract empty traces for the evaluation of the noise standard deviation. It should be provided as [v2, v1], meaning that the noise-only trace of the indirect dimension will be taken at v2 on x_f2, and the noise-only trace of the direct dimension will be taken at v1 on x_f1. Leaving this parameter an None will open the interactive panel, and thus s_reg will be ignored.

guibool

Set it to True to ignore s_reg and n_reg, and use the GUI for the interactive estimation.

Returns:

snr_f1float

The SNR of the indirect dimension

snr_f2float

The SNR of the direct dimension

See also

klassez.anal.noise_std() klassez.anal.snr_gui_2D()

klassez.anal.snr_gui(x, y)

GUI for the evaluation of the Signal to Noise Ratio of a 1D spectrum.

Select “Signal” on the top right corner and drag a region to highlight the reference signal approximate position. The detected point appears as a blue X.

Then, select “Noise”. Drag a signal-free region, i.e. where there is only noise. This will be used for the estimation of the noise standard deviation. The noise level will be highlighted in the figure by two red dashed lines. If these lines do not visually match the noise level, it is most likely there is a signal included in the noise region.

When both the signal and the noise are present, the SNR will be computed. The selection can be refined as many times as one wants, until the figure panel is closed. Close the figure to return the values, and to print the used s_reg and n_reg to be given to klassez.anal.snr().

Parameters:

x1darray

PPM scale of the spectrum

y1darray

Spectrum

Returns:

snrfloat

Estimated signal to noise ratio.

klassez.anal.snr_gui_2D(x_f1, x_f2, yy)

GUI for the evaluation of the Signal to Noise Ratio of a 2D spectrum.

Select “Signal” on the top right corner and drag a rectangle to highlight the reference signal height. The detected point appears as a blue X.

Then, select “Noise”. A red cross-cursor will appear. Find a position where you can extract a signal-free region, i.e. where there is only noise. Double click with the left button of the mouse to extract the projection in that point: they will appear as red traces. These will be used for the estimation of the noise standard deviation.

When both the signal and the noise are present, the SNR will be computed. The selection can be refined as many times as one wants, until the figure panel is closed. Close the figure to return the values, and to print the used s_reg and n_reg to be given to klassez.anal.snr_2D().

Parameters:

x_f11darray

PPM scale of the spectrum for the indirect dimension

x_f21darray

PPM scale of the spectrum for the direct dimension

yy2darray

Spectrum

Returns:

snr_f1float

Estimated signal to noise ratio in the indirect dimension.

snr_f2float

Estimated signal to noise ratio in the direct dimension.

klassez.anal.write_igrl(filename, dic, indirect_scale=None, header=False)

Write a section in a integral report file, which shows the integrated regions and the values of the peaks. It allows

Parameters:

filenamestr

Path to the file to be written

dicdict

Dictionary of integral values. The keys are ‘ppm1:ppm2’ and the associated values are the integrals, as floats or as sequences.

indirect_scale1darray

Scale of the indirect dimension. To be used in the future for fitting

headerbool

If True, adds a “!” starting line to separate fit trials

Returns:

None

See also

klassez.anal.read_igrl()

klassez.anal.integrate()