Pykrev

Added
- spiral_plot function

Changed
- removed zero_ratio argument from formula_definition of element_ratios since it was not used
- updated doc_string for element_counts to include Cl and F
- changed 'Protein' to 'Peptide' and 'Protein-like' to 'Peptide-like' in MSCC and FORM compound class algorithms
- changed 'Phytochemical' to 'Oxy-aromatic phytochemical' in MSCC compound class algorithm
- set nan NP ratios to 0 during MSCC compound class assignment
- updated read_corems to allow removal of peaks with multiple formulae assignments
- updated read_corems to provide verbose output
- updated user_guide and coremswithpykrev example notebooks

Added
- msTuple.filter_bool() method

Changed
- numpy 1.22.0 and networkx 2.5 in requirements
- calculate_mass, aromaticity_index, element_counts, element_ratios and nominal_oxidation_state now support Cl and F (double_bond_equivalent already did)
- user_guide now lists Cl and F support and boolean indexing
- element_ratios no longer includes an optional argument for handling zero ratios (if the denominator is zero, np.nan is now returned)

Added
- read_csv() function
- added a to_csv() method to the msTuple class

Changed
- msTuple.summarise() now gives min and max intensity and std mz in scientific units
- compound_class now supports elemental classification (e.g. CHO, CHON, ...)
- changed the UpSetPlot gude to be up to date with the new version of UpSetPlot.
- made small changes to the bath analysis, PCA and corems with PyKrev guides

Added
- average_mstuple() function

Changed
- mass_spectrum now supports an inverted axis
- mass_spectum now plots onto a number line defined by step size
- added subset and average methods to msTupleDict
- updated the batch analysis and user guide docs to reflect new msTupleDict and mass_spectrum behaviour
- added test for average_mstuple()
- normalise_intensity now supports row or column wise normalisation
- normalise_intensity now supports a range of transformation options, log, power2, and power3
- normalise_intensity now supports centering as a normalisation method

Added
- template function with function docstring in docs
- reaction_network and page_rank functions
- network visualisation notebook in docs
- batch analysis notebook in docs
- msTuple and msTupleDict classes

Changed
- all fuction docstrings edited to match the template function docstring
- all functions now take either msTuple or msTupleDict as primary input, with exception of calculate_mass which retains legacy compatibility
- aromaticity index now allowed to return negative values
- double_bond_equivalent now allowed to return negative values
- read_corems now supports P assignments
- read_corems now checks if C,H,N,O,S,P assignments are present
- read_corems now returns an msTuple
- git ignore now ignores windows batch commands
- changed all relative module imports to specify the exact path
i.e. instead of from ..formula import calculate_mass , we write from ..formula.calculate_mass import calculate_mass
- kendrick mass defect now takes string (as opposed to list) for atom_group
- no more *msTuple as a function input, replaced by msTuple_list, which is clearer and cleaner to work with
- read_batch_formularity now results in an msTuple dictionary instead of a formulaDf
- corems, upsetplot, pca and user guide notebooks to reflect 1.2.0 changes

Deleted
- standardize_formula function as it wasn't used
- unique_formula and missing_formula, as their functionality is just a part of find_intersections and the input syntax was confusing
- unique_plot and missing_plot as they are a type of multi_van_krevelen_plot and the syntax was confusing

Added
- Added __version__ to root directory init.py
- Added read_corems.py function to pykrev/formula
- Added corems_with_pykrev user guide

Changed
- kendrick_mass_defect doesn't require a formula list anymore
- added two extra rounding methods for calculating nominal mass in kendrick mass defect.
'floor' - always round down to nearest integer, and 'ceil' always round up to nearest integer.
- renamed the integer rounding method in kendrick mass defect from 'integer' to 'rint' to avoid confusion.
- removed requirement for formula list from kendrick mass defect plot
- made changes to tests to reflect this
- changed readme to reflect new user guide
- changed read_formularity and read_batch_formularity to automatically exclude all formula with C13 assignments
- changed the user guide to clarify on PyKrev's formula handling, including isotopologues and valid elements.
- changed the input and return order of filter_spectral_interference
- filter_spectral_interference now returns filter_formula as a list, not a numpy array
- monoisotopic masses in calculate_mass made more precise
- monoisotopic masses now allows you to calculate the mass of charged ions and deprotonated ions
- updated the user guide to show how mass error can be calculated using calculate mass
- read_formularity now only returns formula, intensity and mass .Note all functions should input/return in this order.
- normalise_intensity now includes pareto scaling
- addede a test case for calculate_mass to test_formularity_unittest