Skip to contents

Changelog

Source: NEWS.md

eprscope 0.1.14

2025-05-11

Bug Fixes/Critical Updates

  • the entire data frame/table of interactively simulated EPR spectrum within the shinyApp01 (see also the plot_eval_ExpSim_app) can be seamlessly exported either to .csv or to .xlsx formats, where in the previous version only several observations/rows (not the entire data frame) were exported

  • fixed bug in the plot_EPR_Specs where the plot title was not displayed properly (showing a large space between the panel and the title); now, if the user forgets to define the line.colors argument, showing the EPR spectral series (e.g. kinetics), it automatically switches to predefined color vector and the corresponding message is displayed for both the discrete as well as for the continuous color schemes

  • several fixes and/or updates in documentation (functions, including Examples + vignettes)

  • if the B.unit argument, within the eval_sim_EPR_iso equals to different units like B.unit = "G", B.unit = "mT" or B.unit = "T", the intensity scale for the pseudo-Voigt line-shape was different, now it is normalized/standardized to that, corresponding to B.unit = "G", this is especially important for the eval_sim_EPR_isoFit where the initial multiplication intensity coefficient guesses should be comparable (regardless of the B.unit)

  • simulated EPR spectra from different sources can be read by the readEPR_Sim_Spec, taking into account original ASCII data files with several columns (not only two); therefore, now the .csv output of the simulated spectrum corresponding data frame from the shiny plot_eval_ExpSim_app can be combined with the quantify_EPR_Sim_series to quantify rather noisy EPR spectral series and consequently to evaluate the radical kinetics

Updates

  • simplified (several arguments removed) general automatic reading (for the Xenon, WinEpr and Magnetech instrument acquisition toolboxes) for ASCII data/EPR spectral series by the readEPR_Exp_Specs_kin as well as by the readEPR_Exp_Specs_multif functions; additional arguments reduction/simplification in the quantify_EPR_Sim_series and eval_sim_EPR_isoFit functions

  • simplified code for the initial optimization parameter boundaries in the eval_sim_EPR_isoFit

  • package dependencies updated according to new functions/features (see below)

  • README updated by detailed description of package pre-installation setup + R learning resource(s) added

  • linear logarithmic model added to the eval_kinR_Eyring_GHS (TST theory)

  • the functionality vignette/article updated according to the new functions/features (see below)

New Functions/Files/Vignettes

  • when comparing several models/fits (of the same experimental data) a new general ranking function by Akaike and Bayesian information criteria (AIC and BIC) was created: eval_ABIC_forFit, which has been also implemented in the eval_kinR_EPR_modelFit, eval_kinR_Eyring_GHS, eval_sim_EPR_isoFit as well as in the new eval_sim_EPR_isoFit_space (see below) functions

  • for the same purpose (i.e. for comparison of several models/fits and looking for the best one), the new plot_eval_RA_forFit was created, serving as a general diagnostic tool for models/fits based on simple Residual Analysis or analysis of residuals, which correspond to difference between the experimental and predicted values by the model/fit: r (or ε) = yi - ŷi; the function returns several visual diagnostics (plots) together with the standard deviation (sd) of residuals/errors (r/ε); similarly as for the eval_ABIC_forFit, it has been already implemented in the eval_kinR_EPR_modelFit, eval_kinR_Eyring_GHS, eval_sim_EPR_isoFit as well as in the new eval_sim_EPR_isoFit_space (see below)

  • the new eval_sim_EPR_isoFit_space function extends the eval_sim_EPR_isoFit, in order to explore broader range of the initial EPR simulation parameters and to estimate the uncertainties related to optimized ones; additionally, by using the parallel computation, the entire evaluation/iteration process is faster than by using the common sequential method; function returns a .gif animation as well as several plots and data frames in order to show the detailed progress of the fit (corresponding to scanning of the initial parameter hyperspace)

  • in order to demonstrate the essential syntax of R programming/statistical language, applied in the eprscope package, the new firstStep vignette/article was created

eprscope 0.1.13

2025-03-22

Bug Fixes/Critical Updates

  • minimum sum of residual squares (min.rss), coming from the final list of eval_sim_EPR_isoFit, is now identical to that of the calculated from the final data frame df of the column/spectrum Residuals

  • intensity (multiplet) pattern related to just one group of equivalent nuclei is displayed properly when running the eval_sim_EPR_iso + right now the function has no limitations, regarding the number of nuclei (within a group) or their corresponding spin quantum number I → calculation of multiplets (multinomial coefficients) was significantly updated using a recursive function

  • bibliography (.bib) template, when running the create_qmdReport_proj, now possesses the right name inherited from the wd.subdir.name

  • fixed bug in plot_EPR_Specs where the g-value scale was not displayed properly; now, if used either with the plot_theme_NoY_ticks or plot_theme_In_ticks, the back-ticks on the opposite axis are shown as expected

  • several fixes and/or updates in documentation (functions, including Examples + vignettes)

Updates

  • the maximum number of components in quantify_EPR_Sim_series was increased from 6 to 10; however one should be aware of such a high number of radicals when describing the EPR spectrum “envelope” (the sum of all components) because all the components/radicals must mirror the chemical reality of that mixture

  • data frame output from the eval_sim_EPR_isoFit when output.list.forFitSp = TRUE was replaced by the plot/spectrum, depending on the check.fit.plot argument, in order to be ready for the new complex fitting function (currently under development); this is also the reason why the list of suggested packages in the DESCRIPTION has been updated

  • code in several functions → cleaned up in order to be more readable

  • right now, all functions based on optimization of EPR simulation parameters, like optim_for_EPR_fitness, eval_simEPR_isoFit, quantify_EPR_sim_series, possess an option to display messages and progress of the optimization/fitting procedure (including the elapsed time) within the R console, for better interactivity; Examples of those functions have been updated accordingly

  • the Blim argument (defining the magnetic flux density, B region as a vector) was added/updated in several functions which return plots/spectra

New Functions/Files/Vignettes

  • R Shiny application with a simple user interface (UI): plot_eval_ExpSim_app, providing not only interactive visualization of a CW EPR spectrum (and its corresponding instrumental parameters), but also the simulation in isotropic regime; all graphs/spectra and data frames/tables can be also exported to common formats like .csv, .pdf, .png, .jpeg and excel

eprscope 0.1.12

2024-12-05

Bug Fixes/Critical Updates

  • README documentation → fixed/updated, because it did not contain the information about the JDK installation as well as that about the essential R packages + couple of learning or tutorial resources and comments were added

  • several fixes and/or updates in documentation (functions + vignettes)

  • the default intensity multiplication coefficient (parameter) boundaries within the eval_sim_EPR_isoFit function are fixed and correctly set up + extended default boundaries for hyperfine coupling constants (A) added

  • fixed visualization of an EPR spectrum fit within the eval_sim_EPR_isoFit + the initial simulation (see updates) also includes a baseline correction (now it should also work for the consecutive optim.method (vector))

  • increased decimal places of the magnetic flux density (B) and intensity by the reading of ASCII files, using the read_EPR_Specs and related functions

  • fixed issue with the origin argument definition (regarding the spectrometer software labels/names) in reading functions

  • initial qvars for the "(a=1)A + (r=1)R --> [k1] B" model reaction (see also updates) are now correctly set up

  • the minimum sum of residual squares (RSS) value argument was renamed (min.LSQ.sum → min.rss) and unified for all relevant fitting functions

  • tube diameter within the quantify_EPR_Abs is now correctly defined

  • fixed issue with the integral scale within the eval_integ_EPR_Spec

Updates

  • code in several functions → cleaned up in order to be more readable, this also includes renaming of several arguments in those functions + messages and stops added on couple of places, within the code, for the sake of interactivity/awareness of operations

  • reorganization of files and folders related to vignettes

  • code for the integration of EPR spectra (see the eval_integ_EPR_Spec function) was simplified and supplemented with the option to integrate spectra having the [B] = T units, this also applies to the quantify_EPR_Sim_series function

  • the initial simulated EPR spectrum was added to output graphics of the eval_sim_EPR_isoFit value/list (for the case of check.fit.plot = TRUE) in order to compare it with the final/best fit as well as with the experimental EPR spectrum

  • number of letters in kinetic reaction schemes were decreased in order to be more readable + the last scheme (see also bug fixes above) as well as the time duration and resolution were simplified in order to better work with the numeric ODE integration (by the deSolve::ode function)

  • option to select a method for the numeric ODE integration (argument solve.ode.method) was added to both key functions analyzing the radical kinetics

  • added option to select a specific implementation of the pswarm (“particle swarm”) optimization algorithm/method, within the optim_for_EPR_fitness → either “SPSO2007” (for small particle spaces and swarm sizes) or “SPSO2011” (for larger ones) can be set up + internal function updated (not provided as a user option), the number of “informants” (related to particles) was slightly increased by the corresponding exponent, k = 6 (see also documentation for the pso::psoptim)

  • value/output list of the eval_sim_EPR_isoFit was extended to exclusively include vector of the optimized parameters (together with the min.rss) and data frame of the best EPR simulation fit in order to be ready for the new complex fitting function (currently under development)

  • simple residual analysis (plots) added to the eval_kinR_EPR_modelFit output/value list, including the Q-Q (“quantile-quantile”) plot (to check weather residuals are normally distributed) as well as the Residual plot (residuals vs predicted/fitted qvarRs, e.g. concentration) in order to check the kinetic model fit and precision of the optimized kinetic parameters

New Functions/Files/Vignettes

  • create_qmdReport_proj → create file-folder structure of an EPR project to create a reproducible Quarto report (please, also refer to the https://quarto.org/ website)

  • peak picking in EPR/ENDOR spectrum: eval_peakPick_Spec, with the selection of positive or negative intensity peaks for both derivative as well as integrated forms of an EPR/ENDOR spectrum

  • evaluate activation parameters of an elementary radical reaction, obtained by the “Eyring fit” of the experimental k vs T dependence, based on the essential transition state theory (TST) → eval_kinR_Eyring_GHS

eprscope 0.1.11 (1st public release)

2024-08-29

Bug Fixes/Critical Updates

  • fixed reading of spectral time series within the readEPR_Exp_Specs function

  • reading parameter strings from ‘parameter’ files within the readEPR_param_slct function fixed

  • the g-value scale on x-axis within the plotting functions is now properly displayed

  • if the related time(series) unit is converted in relevant data frames, the corresponding column is renamed in order to inherit the new unit (like time_s) accordingly

  • now, the elements in names argument, within the readEPR_Exp_Specs_multif, can also contain characters like c("a","b","c"), not just numbers, e.g. c("250","260","270")

  • fixed visualization of legend labels within the plot_EPR_Specs function

  • natural abundance of interacting nuclei to calculate the intensities within the eval_sim_EPR_iso and related functions/examples

Updates

  • polynomial to fit the distribution of (B1,Bm) within the cavity was extended in quantify_EPR_Abs. Now, the degrees from 5 to 12 are available

  • additionally, the above-corresponding (B1,Bm) theoretical distribution calculation was removed for now, and will be addressed in the future

  • intensity, B, and the time.series column indices arguments, in general reading function readEPR_Exp_Specs, were renamed (using .id) in order to clearly point to headers in col.names

  • magnettech option added to reading functions to read the EPR spectra and parameters from these instruments, additionally, the ‘general’ reading function readEPR_Exp_Specs was updated to read files, from the most common EPR instruments, more easily by adding the auxiliary data.structure argument

  • the optimization algorithms, other than diff-levenmarq, were removed from the eval_kinR_EPR_modelFit for now, additional optimization methods will be implemented later on

  • many function documentations updated + several examples added + code-cleaning (by styler package) done in many functions + README, DESCRIPTION and vignettes updated, accordingly

  • vignette to describe the EPR simulations and fitting was removed, for now (will be added later), however, the important points from that vignette are summarized in the relevant function documentations

  • stoichiometric coefficient notation as well as reaction model scheme for consecutive reactions within the eval_kinR_ODE_model updated/extended

  • option to use Tesla (T) unit, now available in all g-value evaluations

  • line.type option added to all relevant plotting functions

  • calculation of confidence interval by the eval_interval_cnfd_tVec simplified

  • to read the EPR instrumental parameters by readEPR_params_slct, added option to read several parameters simultaneously (vectorized input)

  • customized plot themes, like plot_theme_In_ticks, were simplified, now, there is no need to add opposite axis-ticks by additional theme layer, those function can now also include additional arguments from general theme() function

New Functions/Files/Vignettes

  • perinaphthenyl (PNT) ENDOR spectral data added to package database for documentation examples (in plotting and reading functions)

  • setting up the continuous integrations like R CMD check and pkgdown (in order to build the site) via github actions

  • CONTRIBUTING.md added to package/site

  • new vignette for datasets added

  • new family functions for EPR spectroelectrochemistry, including the function to plot voltammograms and chronoamperograms (plot_ECh_VoC_amperogram) as well as to calculate the transferred charge and the corresponding number of electrons from chronoamperogram (eval_ECh_QNe_chronoamp)

  • new dataset (Triarylamine_radCat_decay_a) as an example of time series experimental data + new datasets coming from magnettech instrument: AcridineDeriv_Irrad_365nm

eprscope 0.1.10

2024-01-24

Bug Fixes/Critical Updates

  • calculation of normalization constant quantify_EPR_Norm_const ⇒ number of points added to calculated norm. constant by unitless receiver gain

  • kinetic models by ODE ⇒ eval_kinR_ODE_model + eval_kinR_EPR_modelFit now consider partial reaction orders and correct formulas for rates ⇒ derivations divided by stoichiometric coefficients according to IUPAC recommendations (see also https://doi.org/10.1021/ed083p510) + additionally, now the kinetic parameters can be optimized not only by the Levenberg-Marquardt algorithm but also by those including within the optim_for_EPR_fitness (in case if the partial reaction orders are considered)

  • eval_kinR_ODE_model can now compare/plot experimental data and model in order to ‘play’ with (manually optimize) partial reaction orders and/or stoichiometric coefficients to fit the experimental data e.g. like integrals or concentrations vs. time. Additionally, the A <– R –> B kinetic model was removed (also from eval_kinR_EPR_modelFit) simply, because it “doubles” the R –> B model. The model can be added later on.

  • several bugs in quantify_EPR_Abs (e.g. like concentration calculation + default temperature definition) were fixed. The function was completely rewritten. Moreover, now the quantification possesses user’s/instrument’s defined polynomial (orders from 6 to 11) fitting of the spatial distribution of B1 and Bm as well as point sample calibration factor as arguments in order to be more flexible. Finally, the theoretical B1 and Bm distribution (see e.g. https://www.sciencedirect.com/science/article/pii/S1090780797912489) can be considered as well in order to estimate the radical concentration.

  • remedy for the Breit-Rabi calculations of energies/frequencies/Bs in simulation => now the g-value for each level is corrected and B are calculated by the fixed-point iterations (see e.g https://doi.org/10.1016/j.jmr.2005.08.013). Simulations are not limited by the number of equivalent nuclei groups anymore (6 could be used up to now). Right now any number of groups maybe used in the actual simulation functions. All simulations are also continuously checked by the package tests/examples.

  • eval_sim_EPR_isoFit now contains the option not only to fit the pseudo-Voight lines but also pure Lorentzian or pure Gaussian onto the experimental EPR spectra

Updates

  • README ⇒ pkg. badges initiated + usage examples added + DESCRIPTION + vignette demonstrating the basic functionality of the package

  • several documentations and examples

  • draw_molecule_by_rcdk ⇒ renamed and focused just on one molecule + added option to place label of the molecule onto an arbitrary position of output image + documentation + examples

  • increased accuracy of ENDOR frequencies in nuclei isotopes_ds dataset

  • couple of variables/arguments (e.g. like B <-> B.val) renamed in order to be consistent throughout the package

  • to select/remove the columns from data frame objects by dplyr the dplyr::all_of() function is now mostly applied in the package code

  • extended tolerance to find half of the max. intensity in order to evaluate FWHM

  • functions to read the EPR spectra like readEPR_Exp_Specs, readEPR_Exp_Specs_kin as well as readEPR_Exp_Specs_multif were simplified and now contain the option to take additional arguments from the essential data.table::fread() which are not predefined in those functions in order to be more flexible upon reading

New Functions/Files/Vignettes

  • added files required to run examples as well as those incl. in vignette(s) ⇒ they are related to EPR spectra of electrochemically generated tetramethyl-phenylenediamine (TMPD) radical cation

  • smoothing of EPR spectra by {npreg} package ⇒ smooth_EPR_Spec_by_npreg which is based on the evaluation of polynomial splines

  • fitting of the experimental EPR spectra by simulations ⇒ eval_sim_EPR_isoFit where several optimization methods can be applied ⇒ all incl. in optim_for_EPR_fitness + baseline correction (either “constant” or “linear” or “quadratic”) is now included in the fitting procedure

  • new tutorial vignette for simulations EPR_Simulations was initiated/set up

  • testing environment by the testthat pkg. set up and initiated + added tests for simulations and conversions

  • added files/spectral data corresponding to an EPR spectrum of aminoxyl radical for examples and testing

eprscope 0.1.9

2023-10-18

Bug Fixes

  • main function to read EPR spectral data (readEPR_Exp_Specs) is now forced to use file argument ⇒ that is data.table::fread(file = path_to_ASC)

  • conversion of time to var within a cyclic change in convert_time2var

  • changed tolerance to find intensity values around 0 in g-factor evaluation (eval_gFactor_Spec) in order to take into account different spectral data resolutions

  • general integration function (eval_integ_EPR_Spec) now includes the proper scaling of sigmoid integrals + fixed Intensity bug in vector output for the integrated form of EPR spectra

  • temperature parameter in instrumental .DSC or .par files is sometimes missing ⇒ now the readEPR_params_tabs checks whether the temperature parameter is present and creates the table accordingly (with or without the temperature)

Updates

  • README, DESCRIPTION and “Introduction”” vignette

  • documentation references set up

  • Code for the absolute quantification of paramag. centers/radicals was simplified within the quantify_EPR_abs function.

  • Function to draw molecules was rewritten to read both .sdf files and smiles characters + it was renamed to draw_molecules_by_rcdk.

  • Reading the simulated spectra (readEPR_Sim_Spec) has now the option to read ASCII data from various sources like ⇒ “easyspin”, “xenon”, “simfonia” (winepr) or “csv”.

  • Instrumental parameters (related to recording of EPR spectra, readEPR_params_tab) can be converted into interactive table based on {DT} package.

  • Both solvents and isotopes datasets were renamed to ..._ds.

  • several documentations and examples

  • The Intensity selection based on different character strings is now replaced by the lineSpecs.form argument throughout the corresponding functions.

  • The simulation section/family was renamed to Simulation and Optimization.

  • Presentation of simulated and experimental spectra (presentEPR_Sim_Spec) has now option to display overlayed spectra.

New Functions

  • read properties of solvents (from the corresponding datasets) ⇒ readEPR_solvent_props

  • convert magnetic flux density (B) values according to input-output units ⇒ convert_B

  • simulation of isotropic EPR spectra in both derivative & integrated spectral forms ⇒ eval_sim_EPR_iso

  • simulation of isotropic EPR spectra corresponding to linear combination of several components evaluated by the latter function ⇒ eval_sim_EPR_iso_combo

  • read instrumental parameters required for the simulations that is now implemented in both sim. functions ⇒ readEPR_params_slct_sim

eprscope 0.1.8

2023-05-15

  • Added functions to evaluate radical kinetics incl. fitting procedures to experimental data (integral intensities).

  • Added basic package data structure and vignettes. Package data now include the isotope database, isotope_db, to gather the basic properties of nuclei mainly for analysis in EPR/ENDOR spectroscopy. For such purpose the new function eval_nu_ENDOR was written. It provides calculation of ENDOR/Larmor frequencies for selected nuclei at specific saturation magnetic flux densities, Bsat.

  • All functions (incl. documentation) are now linked by sections/families.

eprscope 0.1.7

2023-04-25

  • First tagging after main updates in quantify_EPR_sim function. Additionally, the main reading functions, readEPR_... have been completely rewritten.