mocca.peak package

Submodules

mocca.peak.check module

Created on Thu Dec 2 09:16:47 2021

@author: haascp

mocca.peak.check.check_peak(expanded_peak, detector_limit, show_analytics, param=2.5)

Peak checking routine. Returns a checked peak with pure and saturation attributes.

mocca.peak.check.check_peak_purity(peak, show_analytics, param=2.5)

Returns peak purity prediction by performing the described test sequence. Plots and prints infromation about the peak purity prediction.

mocca.peak.check.check_peak_saturation(picked_peak, detector_limit)

Sets peak attribute saturation to either True or False based on if the peak absorbance exceeds detector_limit.

mocca.peak.correct module

Created on Tue Dec 14 15:30:29 2021

@author: haascp

mocca.peak.correct.correct_offset(integrated_peak, istd_peaks, offset)

Creates a corrected peak using internal standard peaks to obtain a retention time offset.

mocca.peak.database module

Created on Fri Nov 26 08:28:12 2021

@author: haascp

class mocca.peak.database.PeakDatabase(peaks: List[ProcessedPeak] | None = None)

Bases: object

Database class to store and organize peaks of HPLC-DAD data.

increment_unknown_counter()

Increments the unknown counter by one.

insert_peak(new_peak)

Inserts a peak in the database given it is of the processed peak type. If a peak with the same borders in the same dataset already exists, it will be overwritten.

update_unknown_counter()

Updates the unknown counter which gives the highest trailing number in the peak database of a peak assigned with a compound_id starting with the string literal ‘unknown’.

mocca.peak.expand module

Created on Thu Dec 2 09:16:47 2021

@author: haascp

mocca.peak.expand.expand_peak(picked_peak, absorbance_threshold)

Expands peak boundaries to those actually in the data. It keeps expanding them until the absorbance falls below one twentieth of the given absorbance threshold. Returns a picked peak with modified peak boundaries (left, right).

mocca.peak.integrate module

Created on Tue Dec 7 10:57:23 2021

@author: haascp

mocca.peak.integrate.integrate_peak(checked_peak)

Integrates the peak. Returns an integrated peak with the integral attribute set.

mocca.peak.match module

Created on Wed Dec 1 12:06:57 2021

@author: haascp

mocca.peak.match.get_filtered_similarity_dicts(peak, component_db, spectrum_correl_coef_thresh, relative_distance_thresh, print_out=False)

Filters the list of similarity dictionaries with regard to the given thresholds. Return possible matches which have a spectral correlation coefficient higher than the given threshold and a relative distance between the peak maxima lower than the given threshold.

mocca.peak.match.get_relative_distance(peak, component)

Returns the distance of an offset-corrected peak maximum and a component maximum relative to the length of the time vector.

mocca.peak.match.get_similarity_dicts(peak, component_db, relative_distance_thresh)

Returns a sorted list of dictionaries. For each component in the given database, similarity values to the given peak are stored.

mocca.peak.match.get_spectrum_correl_coef(peak, component)

Returns the correlation coefficient of the average peak spectrum and the spectrum of the component.

mocca.peak.match.match_peak(corrected_peak, component_db, spectrum_correl_coef_thresh, relative_distance_thresh, print_similarity_dicts=False)

Routine to assign possible matches to a returned preprocessed peak.

mocca.peak.match.update_matches(peak, new_matches)

Updates the matches of a given peak.

mocca.peak.models module

class mocca.peak.models.BasePeak(left: int, right: int, maximum: int, offset: int)

Bases: object

Base peak class.

left: int

maximum: int

offset: int

right: int

class mocca.peak.models.CheckedPeak(left: int, right: int, maximum: int, offset: int, dataset: CompoundData, idx: int, saturation: bool, pure: bool)

Bases: PickedPeak

Class for peaks checked with regard to saturation and purity.

pure: bool

saturation: bool

class mocca.peak.models.CorrectedPeak(left: int, right: int, maximum: int, offset: int, dataset: CompoundData, idx: int, saturation: bool, pure: bool, integral: float, istd: List[IstdPeak])

Bases: IntegratedPeak

Class for peaks with added retention time offset. From this class on, retention times in the peaks are already corrected. This means, that accessing data from the dataset attribute require prior un-offsetting.

istd: List[IstdPeak]

class mocca.peak.models.IntegratedPeak(left: int, right: int, maximum: int, offset: int, dataset: CompoundData, idx: int, saturation: bool, pure: bool, integral: float)

Bases: CheckedPeak

Class for integrated peaks.

integral: float

class mocca.peak.models.IstdPeak(left: int, right: int, maximum: int, dataset: CompoundData, integral: float, offset: int, compound_id: str, concentration: float)

Bases: object

Class for istd peaks to be added to the peak classes below.

compound_id: str

concentration: float

dataset: CompoundData

integral: float

left: int

maximum: int

offset: int

right: int

class mocca.peak.models.PickedPeak(left: int, right: int, maximum: int, offset: int, dataset: CompoundData, idx: int)

Bases: BasePeak

Class for picked peaks out of DAD data. Also valid for expanded peaks.

dataset: CompoundData

idx: int

class mocca.peak.models.PreprocessedPeak(left: int, right: int, maximum: int, offset: int, dataset: CompoundData, idx: int, saturation: bool, pure: bool, integral: float, istd: List[IstdPeak], matches: List[dict])

Bases: CorrectedPeak

Class for preprocessed peaks containing a list of possible component matches in the attribute compound_id.

matches: List[dict]

class mocca.peak.models.ProcessedPeak(left: int, right: int, maximum: int, offset: int, dataset: CompoundData, idx: int, saturation: bool, pure: bool, integral: float, istd: List[IstdPeak] | None = None, compound_id: str | None = None, concentration: float | None = None, is_compound: bool = False)

Bases: object

Class of fully processed peaks ready to be put in the peak database.

compound_id: str | None = None

concentration: float | None = None

dataset: CompoundData

idx: int

integral: float

is_compound: bool = False

istd: List[IstdPeak] = None

left: int

maximum: int

offset: int

pure: bool

right: int

saturation: bool

mocca.peak.process module

Created on Fri Dec 17 11:03:08 2021

@author: haascp

mocca.peak.process.process_peak(peak, compound, is_compound=False)

Creates a processed peak by addding compound information to it.

mocca.peak.purity_funcs module

Created on Tue Nov 23 15:55:25 2021

@author: haascp

mocca.peak.purity_funcs.get_agilent_thresholds(peak_data, max_loc, noise_variance, param=2.5)

Returns the thresholds calculated by the Agilent purity algorithm.

mocca.peak.purity_funcs.get_correls(peak_data, max_loc)

Get a list with correlation coefficients of UV-Vis spectra at every timepoint with reference to the UV-Vis spectrum at maximum absorbance.

mocca.peak.purity_funcs.get_max_loc(peak_data)

Returns the maximum location of the given peak data.

mocca.peak.purity_funcs.get_noise_variance(peak)

Filters dataset with only timepoints whose max absorbance at any wavelength is below 1% of max absorbance. Returns the average of the variance over all wavelengths.

mocca.peak.purity_funcs.get_pca_explained_variance(peak_data)

Calculates the ration of explained variance by the first principal component of the devonvoluted peak data.

mocca.peak.purity_funcs.get_purity_value_agilent(peak_data, correls, agilent_thresholds)

Uses Agilent’s peak purity algorithm to predict purity of peak. Param gives strictness of test (original was 0.5, which is more strict)

mocca.peak.purity_funcs.get_trimmed_peak_data(peak)

Returns peak data trimmed with cut edges of the peak to 5% of max absorbance to avoid noise artifacts.

mocca.peak.purity_funcs.predict_purity_unimodal(correls)

Checks for unimodality of a peak by an averaging filter of length 3 on the correlation vector to the maximum https://stackoverflow.com/questions/14313510/how-to-calculate-rolling-moving-average-using-numpy-scipy

mocca.peak.quantify module

Created on Tue Jan 4 15:49:54 2022

@author: haascp

mocca.peak.quantify.quantify_peak(peak, quant_comp_db, quali_comp_db)

Takes a peak with an integral and quantifies it by calculating the corresponding concentration if the compound is present in the quantification database.

mocca.peak.resolve_impure module

Created on Fri Jan 14 08:59:36 2022

@author: haascp

mocca.peak.resolve_impure.create_parafac_peak(comp_i, parafac_model)

Return synthetic PARAFAC peaks created from the PARAFAC decomposition results.

mocca.peak.resolve_impure.create_pure_peak(impure_peak)

Takes an impure peak and returns its copy with the pure attribute True.

mocca.peak.resolve_impure.get_parafac_data_shift(iter_offset)

If the iteration offset is larger than zero, the impure peak was shifted and must therefore be shifted back in the resulting PARAFAC data. If the iteration shift was negative, the pure signals were shifted and nothing has to be done.

mocca.peak.utils module

Created on Thu Dec 2 10:00:18 2021

@author: haascp

mocca.peak.utils.average_peak_spectrum(peak)

Calculates mean spectrum over peak from left to right border.

mocca.peak.utils.get_peak_data(peak)

Returns absorbance data from the left to the right border of the peak for all wavelengths. If the peak was offset-corrected, the left and right border are un-offset in order to access the correct data.

mocca.peak.utils.get_retention_times(peak)

Returns left and right borders as well as maximum of a peak as retention times.

mocca.peak.utils.is_unimodal(L, high_val_threshold=inf)

Checks if a list is unimodal (for use in peak purity).

Parameters:

• L (list) – A list to test unimodality for

• high_val_threshold (numeric, optional) – If set, then values above high_val_threshold will not be counted in unimodality testing. Default is np.inf (i.e. this threshold is not used).

Returns:

True if the list is unimodal ignoring high values; False otherwise.

Return type:

TYPE boolean

Module contents