Changes

* Support for R1rhoR_1ρ-type data.

* The [[R2eff]] model - used to determine the R2eff R_2eff or R1rho R_1ρ values and errors required as the base data for all other models.

* The [[LM63]] SQ CPMG-type analytic model - the original Luz and Meiboom 1963 2-site fast exchange equation with parameters {R20R₂⁰, ..., phi_exφ_ex, kexk_ex} [Luz and Meiboom 1963].* The [[LM63 3-site]] SQ CPMG-type analytic model - the original Luz and Meiboom 1963 3-site fast exchange equation with parameters {R20R₂⁰, ..., phi_exBφ_ex,B, k_B, phi_exCφ_ex,C, k_Ck_C} [Luz and Meiboom 1963].* The [[CR72]] SQ CPMG-type analytic model - the reduced Carver and Richards 1972 2-site equation for most time scales whereby the simplification R20A R_2A⁰ = R20B R_2B⁰ is assumed with the parameters {R20R₂⁰, ..., pAp_A, dwδω, kexk_ex} [Carver and Richards 1972].* The [[CR72 full]] SQ CPMG-type analytic model - the full Carver and Richards 1972 2-site equation for most time scales with parameters {R20AR_2A⁰, R20BR_2B⁰, ..., pAp_A, dwδω, kexk_ex} [Carver and Richards 1972].* The [[IT99]] SQ CPMG-type analytic model - the Ishima and Torchia 1999 2-site model for all time scales with pA p_A ≫ pB p_B and with parameters {R20R₂⁰, ..., phi_exφ_ex, pAp_A.dwδω^2, kexk_ex} [Ishima and Torchia 1999].* The [[TSMFK01]] SQ CPMG-type analytic model - the Tollinger et al., 2001 2-site very-slow exchange model for time scales within range of microsecond to second time scale with parameters are {R20AR_2A⁰, ..., dwδω, k_ABk_AB} [Tollinger et al., 2001].* The [[NS CPMG 2-site expanded]] SQ CPMG-type numeric model - A model for 2-site exchange expanded using Maple by Nikolai Skrynnikov (Tollinger et al., 2001) with the parameters {R20R₂⁰, ..., pAp_A, dwδω, kexk_ex}.* The [[NS CPMG 2-site 3D]] SQ CPMG-type numeric model - the reduced model for 2-site exchange using 3D magnetisation vectors whereby the simplification R20A R_2A⁰ = R20B R_2B⁰ is assumed with the parameters {R20R₂⁰, ..., pAp_A, dwδω, kexk_ex}.* The [[NS CPMG 2-site 3D full]] SQ CPMG-type numeric model - the full model for 2-site exchange using 3D magnetisation vectors with parameters {R20AR_2A⁰, R20BR_2B⁰, ..., pAp_A, dwδω, kexk_ex}.* The [[NS CPMG 2-site star]] SQ CPMG-type numeric model - the reduced model for 2-site exchange using complex conjugate matrices whereby the simplification R20A R_2A⁰ = R20B R_2B⁰ is assumed with the parameters {R20R₂⁰, ..., pAp_A, dwδω, kexk_ex}.* The [[NS CPMG 2-site star full]] SQ CPMG-type numeric model - the full model for 2-site exchange using complex conjugate matrices with parameters {R20AR_2A⁰, R20BR_2B⁰, ..., pAp_A, dwδω, kexk_ex}.* The [[M61]] R1rhoR_1ρ-type analytic model - the Meiboom 1961 2-site fast exchange equation for on-resonance data with parameters {R1rhoR_1ρ', ..., phi_exφ_ex, kexk_ex} [Meiboom 1961].* The [[M61 skew]] R1rhoR_1ρ-type analytic model - the Meiboom 1961 2-site equation for all time scales with pA p_A ≫ pB p_B and with parameters {R1rhoR_1ρ', ..., pAp_A, dwδω, kexk_ex} [Meiboom 1961].* The [[DPL94]] R1rhoR_1ρ-type analytic model - the Davis et al., 1994 2-site fast exchange equation extending the 'M61' model for off-resonance data with parameters {R1rhoR_1ρ', ..., phi_exφ_ex, kexk_ex} [Davis et al., 1994].* The [[TP02]] R1rhoR_1ρ-type analytic model - the Trott and Palmer 2002 2-site equation for all time scales with pA p_A ≫ pB p_B and with parameters {R1rhoR_1ρ', ..., pAp_A, dwδω, kexk_ex} [Trott and Palmer 2002].* The [[TAP03]] R1rhoR_1ρ-type analytic model - the Trott et al., 2003 off-resonance 2-site equation for all time scales with the weak condition pA p_A ≫ pB p_B and with parameters {R1rhoR_1ρ', ..., pAp_A, dwδω, kexk_ex} [Trott et al., 2003].* The [[MP05]] R1rhoR_1ρ-type analytic model - the Miloushev and Palmer 2005 off-resonance 2-site equation for all time scales with parameters {R1rhoR_1ρ', ..., pAp_A, dwδω, kexk_ex} [Miloushev and Palmer 2005].* The [[NS R1rho 2-site]] R1rho R_1ρ numeric model - the model for 2-site exchange using 3D magnetisation vectors with the parameters {R1rhoR_1ρ', ..., pAp_A, dwδω, kexk_ex}.* The [[MQ CR72]] MMQ-type analytic model - the Carver and Richards 1972 2-site model for most time scales expanded for MMQ CPMG data by Korzhnev et al., 2004 with the parameters {R20R₂⁰, ..., pAp_A, dwδω, dwHδω_H, kexk_ex}.* The [[MMQ 2-site]] MMQ-type numeric model - the model for 2-site exchange whereby the simplification R20A R_2A⁰ = R20B R_2B⁰ is assumed with the parameters {R20R₂⁰, ..., pAp_A, dwδω, dwHδω_H, kexk_ex}.

* Introduced a RelaxError when chosing 'r1rho' as experiment type as this won't be implemented now. Efforts will be concentrated on the CPMG code first, then on the R1rho R_1ρ code.

* Started to implement a function for calculating the effective transversal relaxation rate (R2effR_2eff). This follows a thread at https://mail.gna.org/public/relax-devel/2009-01/msg00067.html.

* Started to implement the scaling matrix for scaling the 'R2effR_2eff' values. This might change in the future as other possible curve fitting parameters ('R2', 'Rex', 'kexk_ex', 'R2AR_2A⁰', 'kA', 'dwδω') might need some scaling.

* Started to implement a function for reading 'R2effR_2eff' values directly. This is as proposed in a post at https://mail.gna.org/public/relax-devel/2009-01/msg00020.html. The function does not contain code yet.

* Corrected the way the scaling matrix is assembled. This is as proposed by Ed in a post at https://mail.gna.org/public/relax-devel/2009-01/msg00079.html. The scaling values are now based on the default values for the different parameters which were slightly modified. The only parameter for which the average is still used (as for intensities in the 'relax_fit.py' code) is 'R2effR_2eff'.

* Started to make changes for multiple field relaxation dispersion analysis. This seems necessary, so maybe we should not support single field analysis of relaxation dispersion at all. -> Kovrigin et al. (2006) JMagRes, 180: 93-104. The changes made here are only a first draft and may not work. In particular, maybe the spectrum.read_intensities(), relax_disp.cpmg_frq(), spectrum.replicated(), spectrum.error_analysis(), and deselect.read() functions will need to know the magnetic field to which the particular dataset is associated... In fact, the different datasets should be input first and their R2eff R_2eff calculated independently. In a second step, the actual relaxation dispersion curve fitting should be made with all data.

* Removed the obsolete function 'relax_disp.r2eff_read()'. R2eff R_2eff values will be read directly by 'relax_data.read()'.

* A small fix to the system test. However, is this fix the solution or is there something wrong with the reading of data (such as 'R2effR_2eff') by relax_data.read()? Shouldn't the data, for example 'R2effR_2eff', be available in 'cdp.mol[0].res[0].spin[0].R2eff_val[0]' or 'cdp.mol[0].res[0].spin[0].R2eff[0]' for the 1st spin of the 1st residue in the 1st molecule?

* Created the relax_disp.spin_lock_field user function. This is used to set the spin-lock field strength of a given R1rho R_1ρ spectrum.

* Fixes for the dispersion specific analysis separating R2eff R_2eff from R2R₂. There is one R2eff R_2eff parameter per exponential curve, but only one R2 R₂ per model. The code now better handles this.* The dispersion specific methods now handle one R2eff R_2eff and I0 I₀ parameter per exponential curve.

* The dispersion specific _assemble_param_vector() method now handles multiple R2eff R_2eff and I0 I₀ values. These spin structures are dictionaries holding multiple values.

* Expanded the relax_disp.select_model user function documentation to cover R2eff R_2eff and I0I₀. These parameters and how they are optimised are now better described.

* Added some synthetic data to test the 'exp_fit' relaxation dispersion model fitting. These are just basic synthetic exponential curves. R2eff R_2eff and I0 I₀ should be very easy to find.

* Fixes for the dispersion specific _assemble_param_vector() method. The R2eff R_2eff and I0 I₀ spin dictionary structures are now checked for their keys before pulling the value out.

* Started to redesign how R2eff R_2eff is handled in the relaxation dispersion analysis. Instead of being part of the optimisation of the dispersion model, it will itself be the model named 'R2eff' (converted from the 'exp_fit' model). This model will either use the calc user function to determine R2eff R_2eff when the fixed relaxation period experiment is selected, or fit exponential curves using the relax_fit C module for the variable relaxation period experiments. The R2eff R_2eff values will then be copied over for each dispersion model in the auto-analysis using the value.copy user function.

* Expanded the new analysis wizard in the GUI to accommodate the relaxation dispersion auto-analysis. Now the buttons are a matrix of 4x2 with the NOE, R1R₁, R2R₂, and model-free analyses at the top and two new blank buttons have been added to the bottom. One will be used for the dispersion analysis.

* Created the initial version of the relaxation dispersion auto-analysis GUI frame. This is built from copying lots of code from the NOE, R1R₁, and R2 R₂ analyses. The dispersion specific parts will be added later.

* Elimination of the relax_disp.calc_r2eff user function. This user function, which is non-functional anyway, is not needed. The calculation of the R2eff R_2eff values will occur with the optimisation of the 'R2eff' model (with a call to the calc user function for the fixed time period experiment types) so direct calculation through a specific user function is not needed.

* Redesign of the relaxation dispersion models in the relax_disp.select_model user function front-end. The models have been renamed and better defined based on the experiment type (CPMG or R1rhoR_1ρ).* The relaxation dispersion scaling matrix assembly now uses lib.mathematics.round_to_next_order(). This allows the printed out I0 I₀ values for the optimisation of the exponential curves to be easier to scale back to the real value.* The Relax_disp.test_hansen_cpmg_data_fast_2site system test now uses the R2eff model. The equivalent slow exchange system test also uses the model. This model will be used to find the R2eff R_2eff values from the fixed relaxation time period data.

* Started to implement the relaxation dispersion analysis specific calculate() method. This will be used to calculate the R2effR<sub>2eff</R1rho sub>/R_1&rho; values for the fixed relaxation time period data types and is equivalent to Sebastien Morin's relax_disp.calc_r2eff user function which was deleted (see http://thread.gmane.org/gmane.science.nmr.relax.scm/17336).

* Created the lib.dispersion.calc_two_point_r2eff() function. This is for calculating the R2effR<sub>2eff</R1rho sub>/R_1&rho; value for the fixed relaxation time data.

* Completed the relaxation dispersion calculate() method. This allows the R2effR<sub>2eff</R1rho sub>/R_1&rho; values to be calculated for the fixed relaxation time period experiments through the calc user function.

* Changed the phi_ex &phi;_ex parameter in the LM63 model back to rex.* Changed the Grace string for the rex parameter to be phi_ex&phi;_ex.* Converted all of the specific_analyses.relax_disp.parameters module to handle different models. The R2eff R_2eff and I0 I₀ parameters are now only part of the 'R2eff' model and all other standard parameters belong to all of the other models.

* Started to redesign the relaxation dispersion target function class. The input data is now expected to be R2effR<sub>2eff</R1rho sub>/R_1&rho; data and all mentions of exponential curves have been eliminated. The func_exp_fit() target function has been deleted as it is not used - as now the _minimise_r2eff() method in the dispersion specific analysis class is used instead. And the func_fast_2site() target function has been renamed to func_LM63().* Redesigned the optimisation code of the dispersion analysis specific class for the new target functions. This includes the assembling of R2effR_2eff/R<sub>1&rho;</R1rho sub> values instead of peak heights, and a number of small fixes.

* Shifted the R2effR<sub>2eff</R1rho sub>/R_1&rho; value and error assembly into specific_analyses.relax_disp.disp_data. This is in the new return_r2eff_arrays() function. The code has also been debugged and made functional.

* Major redesign of the relaxation dispersion data model in the relax data store. The data model now revolves around the three concepts of the spectrometer frequency, the dispersion points, and the relaxation times. Peak intensity data is now handled through averaging using the new specific_analyses.relax_disp.disp_data.average_intensity() function. R2effR<sub>2eff</R1rho sub>/R_1&rho; values are now referenced by a key generated from the spectrometer frequency and nu_CPMG frequency or nu1 spin-lock field strength. All of the specific_analyses.relax_disp package has been standardised around these concepts. This simplifies all of the modules of the package and removes a large number of latent bugs.

* Fixes for the LM63 dispersion CPMG model. The 'r2' model parameter is now an array as there is one R2 R₂ value per magnetic field strength. And the 'rex' parameter has been renamed to 'phi_ex&phi;_ex' and is scaled quadratically with the field strength within the optimisation target function.

* Created the Relax_disp.test_hansen_cpmgfit_input system test. This is for checking the operation of the relax_disp.cpmgfit_input user function conversion of Flemming Hansen's CPMG R2eff R_2eff values into input files for CPMGFit. A relax state file containing the results of an analysis of an 'R2eff' model analysis of the truncated data has been added to the test suite data to speed up the test and to check the loading of dispersion state files.

* The specific_analyses.relax_disp.disp_data.loop_frq() function can now handle missing data. This allows the loop to yield a single value of None when the spectrometer information has not been loaded and enables R1rho R_1&rho; analyses at a single field strength.* Fix for the LM63 dispersion model target function - the scaled Phi_ex &phi;_ex value is now used for the R2R₂.* Fixes for the relaxation dispersion auto-analysis for the LM63 model. The Rex parameter is now the phi_ex &phi;_ex parameter.

* Massively increased the precision of the R2eff R_2eff error analysis. The hard-coded simulation number variable is now set to 100000. This appears to be necessary for reliably reproducing results in the subsequent dispersion models.

* Created a script to calculate the R2eff R_2eff rate errors extremely precisely for Hansen's CPMG data. This uses 1 million Bootstrap simulations for calculating the errors. The 'r2eff_values.bz2' is saved after deleting the spin specific r2eff_sim structures so that the file drops from 388 Mb in size to 7.3 kb.

* Updated the NESSY results for Flemming Hansen's CPMG data for the R2eff R_2eff values with high precision errors. A file containing the log or printouts from NESSY has been added for reference.

* Large simplification of the Relax_disp system tests using Hansen's CPMG data. Instead of calculating the R2eff R_2eff values in the test, these are read from the high error precision results files in test_suite/shared_data/dispersion/Hansen. This allows the model parameters to be consistently found and to be identical between different runs of the test.

* An upper limit of 200 rad/s has been added to the linear constraints for the R2 R₂ dispersion parameters.

* Implemented a basic graph for the relax_disp.plot_disp_curves user function. This simply plots out the nu_CPMG value or spin-lock field verses the R2effR_2eff/R<sub>1&rho;</R1rho sub> values from the experiment. The graph of the back calculated R2effR_2eff/R<sub>1&rho;</R1rho sub> values from the model fit is still to be added.* Fix for the linear constraints for the 'R2eff' model. The A and b matrices are no longer set to None, as this kills the auto-analysis or any analysis when constraints are turned on. Now the constraints 0 <= R2eff R_2eff <= 200 and I0 I₀ >= 0 are used.

* Fix for the phi_ex &phi;_ex dispersion parameter scaling - the scaling was in the wrong direction.* Added a directory of relax results for the truncated high-precision Hansen CPMG R2eff R_2eff values. This uses the base_pipe.bz2 and r2eff_pipe.bz2 files with the high-precision R2eff R_2eff errors, and hence can be compared to the NESSY and CPMGFit results.* Added the residue :71 results to the lm63_comparison file. This is a summary of the optimisation results using the high-precision R2eff R_2eff error results for the different dispersion softwares.

* Added the ShereKhan results for the high-precision R2eff R_2eff data for Hansen's CPMG data.

* Modified the dispersion calculate() method for the R2eff R_2eff values to use the analytic equation. For the R2effR_2eff/R<sub>1&rho;</R1rho sub> values calculated for the fixed time period dispersion experiments via the calc user function, the very slow and tedious bootstrapping approach has been replaced by the very quick direct error calculation. The two techniques produce the same results as the bootstrap simulation number approaches infinity.

* The relax_disp.relax_time page is now always shown in the peak intensity wizard for the dispersion GUI. This number is needed for the fixed time period experiments as well to calculated the R2effR<sub>2eff</R1rho sub>/R_1&rho; values and errors.

* Simplified the pA p_A >= pB p_B constraint in the dispersion linear_constraints() function.

* Improvements for the phi_ex &phi;_ex and dw &delta;&omega; relaxation dispersion model parameters. These are now stored with the units of ppm^2 and ppm respectively. The conversion to (rad/s)^2 and rad/s units respectively now is spin specific, allowing mixed spin types (1H, 13C, 15N, etc.) to be analysed simultaneously.* Updated the relax results for Hansen's CPMG data for the recent phi_ex &phi;_ex and dw &delta;&omega; changes.

* Fix for the grid search setup for the pA p_A dispersion parameter. As pA p_A > p_B pB, then the region from 0.0 to 0.5 does not need to be searched.* The back calculated R2eff R_2eff values are now stored for dispersion analysis after minimisation.

* Updated the NESSY log file for its improved printouts. These printouts allow the R20 R₂⁰ values to be accessed.

* Updated the relax_disp.select_model user function docstring for the R2eff R_2eff error analysis. This properly describes how the R2effR_2eff/R<sub>1&rho;</R1rho sub> errors are calculated for the fixed relaxation time period experiments.

* Expanded the number of model list variables in specific_analyses.relax_disp.variables. This is to include lists which are specific to CPMG-type and R1rhoR_1&rho;-type experiments.* Added the new 'M61' model to the specific_analyses.relax_disp.variables module. This is the Meiboom 1961 model for 2-site fast exchange for R1rhoR_1&rho;-type experiments.* Added the M61 model to the relax_disp.select_model user function frontend. This is the Meiboom 1961 model for 2-site fast exchange for R1rhoR_1&rho;-type experiments.* Added the M61 model equations to the relax library. This is for the Meiboom 1961 2-site fast exchange model for R1rhoR_1&rho;-type experiments.* Created the M61 2-site fast exchange dispersion model target function. This is for the Meiboom 1961 model for 2-site fast exchange for R1rhoR_1&rho;-type experiments. The code for the func_M61() method was copied without modification from the func_LM63() method.* Added support for the R1rhoR_1&rho;-type experiments to the relaxation dispersion analysis in the GUI. This involves using a different model list for these experiments compared to the CPMG-type experiments.

* Fix for the Relax_disp.test_hansen_cpmg_data_auto_analysis system test. The correct model list is now being used - the R1rhoR_1&rho;-type experiments should not be included.

* Added support for the M61 model to the relax_disp.select_model user function back end. This is for the Meiboom 1961 2-site fast exchange model for R1rhoR_1&rho;-type experiments.

* Added some synthetic on-resonance R1rho R_1&rho; data to the test suite. This is in the form of Sparky peak list files containing two spin systems.* Expanded the synthetic on-resonance R1rho R_1&rho; test suite data. The data now consists of a full set of dispersion curves for the M61 model.* Added a reference to the synthetic on-resonance R1rho R_1&rho; test suite data. The first ncyc1 data point now has a relaxation time period of zero, hence it can be used as the reference for a fixed time period experiment.

* Variable renaming in the lib.dispersion.m61 module. The variable names are now more suited to R1rhoR_1&rho;-type data, rather than CPMG-type data.

* Created the Relax_disp.test_r1rho_on_res_fixed_time_m61 system test. This checks the R1rhoR_1&rho;-type experiment with a fixed relaxation time period using the 'R2eff' and 'M61' models. It uses the auto-analysis for this, and the 'r1rho_on_res' synthetic relaxation data.

* Improvements to the relax_disp.plot_disp_curves user function. The back-calculated R2effR<sub>2eff</R1rho sub>/R_1&rho; values are now included in the plot as separate sets. In addition, the residuals have also been added to allow for a visual statistical comparison.* More improvements to the relax_disp.plot_disp_curves user function. The data sets now have labels, and the residuals have errors set to those of the R2effR_2eff/R<sub>1&rho;</R1rho sub> data.

* Fix for the r2eff_calc.py script for calculating R2eff R_2eff values from Flemming Hansen's CPMG data.* Added a check to the dispersion specific minimise() function for the spectrometer field strength. This is essential in all dispersion models to convert between ppm and rad/s units, or ppm^2 and (rad/s)^2 for the phi_ex &phi;_ex parameter.

* Added the 'M61 skew' model to the specific_analyses.relax_disp.variables module. This is the Meiboom 1961 model for skewed populations (pA p_A &#x226B; pBp_B). This commit follows step 1 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the 'M61 skew' model to the relax_disp.select_model user function frontend. This is the Meiboom 1961 model for skewed populations (pA p_A &#x226B; pBp_B). This commit follows step 2 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).

* Added the M61 skew model equations to the relax library. This is the Meiboom 1961 on-resonance 2-site model for skewed populations (pA p_A &#x226B; pBp_B). This commit follows step 3 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Created the M61 skew model target function. This is the Meiboom 1961 on-resonance 2-site model for skewed populations (pA p_A &#x226B; pBp_B). This commit follows step 4 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added support for the skewed condition (pA p_A &#x226B; pBp_B) to the specific_analyses.relax_disp.parameters module. This is currently done by constraining pA p_A to be greater than 0.85.* Added support for the M61 skew model to the relax_disp.select_model user function back end. This is the Meiboom 1961 on-resonance 2-site model for skewed populations (pA p_A &#x226B; pBp_B). This commit follows step 6 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).

* Created test data for the 'M61 skew' R1rho R_1&rho; model. This is the Meiboom 1961 on-resonance 2-site model for skewed populations (pA p_A &#x226B; pBp_B). This commit follows step 7 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Created the Relax_disp.test_r1rho_on_res_fixed_time_m61b system test. This is for the Meiboom 1961 on-resonance 2-site model for skewed populations (pA p_A &#x226B; pBp_B). This commit follows step 7 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).

* Updates for the pA p_A dispersion parameter optimisation constraints. The parameter is now limited to be between pB p_B and 1. In the case of the limit pA p_A &#x226B; pBp_B, then instead the constraint is between 0.85 and 1.

* Fixes for the grid search for the 'M61 skew' dispersion model. The pA p_A parameter search is now between 0.85 and 1.* Fixes for the func_M61b() dispersion target function. This is the Meiboom 1961 on-resonance 2-site model for skewed populations (pA p_A &#x226B; pBp_B).* Small changes to the r1rho_on_res_m61b dispersion test data. One R20 R₂⁰ rate has been increased.* Completed the lib.dispersion.m61.r1rho_M61() function. Now the R1 R₁ relaxation rate and rotating frame tilt angle are correctly handled. This is not used in the target functions as support for the R1 R₁ and offset is not yet implemented.* Added the 'DPL94' model to the specific_analyses.relax_disp.variables module. This is the David, Perlman and London 1994 R1rho R_1&rho; 2-site fast exchange model. This commit follows step 1 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the 'DPL94' model to the relax_disp.select_model user function frontend. This is the David, Perlman and London 1994 R1rho R_1&rho; 2-site fast exchange model. This commit follows step 2 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).

* Added the DPL94 model equations to the relax library. This is the David, Perlman and London 1994 R1rho R_1&rho; 2-site fast exchange model. This commit follows step 3 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Created the DPL94 model target function. This is the David, Perlman and London 1994 R1rho R_1&rho; 2-site fast exchange model. This commit follows step 4 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added support for the DPL94 model to the relax_disp.select_model user function back end. This is the David, Perlman and London 1994 R1rho R_1&rho; 2-site fast exchange model. This commit follows step 6 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Created the Relax_disp.test_r1rho_on_res_fixed_time_dpl94 system test. This is the David, Perlman and London 1994 R1rho R_1&rho; 2-site fast exchange model. This commit follows step 7 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the 'IT99' model to the specific_analyses.relax_disp.variables module. This is the Ishima and Torchia 1999 2-site model for all timescales with pA p_A &#x226B; pBp_B. This commit follows step 1 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the 'IT99' model to the relax_disp.select_model user function frontend. This is the Ishima and Torchia 1999 2-site model for all timescales with pA p_A &#x226B; pBp_B. This commit follows step 2 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Updated the model lists of the dispersion analyses GUI element. This adds the IT99 CPMG-type model and the DPL94 and M61B R1rhoR_1&rho;-type models.* Fixes for the IT99 model description in the relax_disp.select_model user function. This is the Ishima and Torchia 1999 2-site model for all timescales with pA p_A &#x226B; pBp_B. This commit follows step 2 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the IT99 model equations to the relax library. This is the Ishima and Torchia 1999 2-site model for all timescales with pA p_A &#x226B; pBp_B. This commit follows step 3 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the it99 module to the lib.dispersion package __all__ list. This is the Ishima and Torchia 1999 2-site model for all timescales with pA p_A &#x226B; pBp_B. This commit follows step 3 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Created the IT99 model target function. This is the Ishima and Torchia 1999 2-site model for all timescales with pA p_A &#x226B; pBp_B. This commit follows step 4 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Fix for the lib.dispersion.it99 module. This is the Ishima and Torchia 1999 2-site model for all timescales with pA p_A &#x226B; pBp_B. This commit follows step 3 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the support for the pAp_A.dw&delta;&omega;^2 parameter 'padw2' to the dispersion specific analysis. This is needed for the Ishima and Torchia 1999 2-site model for all timescales with pA p_A &#x226B; pBp_B. This commit follows step 5 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added support for the IT99 model to the relax_disp.select_model user function back end. This is the Ishima and Torchia 1999 2-site model for all timescales with pA p_A &#x226B; pBp_B. This commit follows step 6 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the support for the tex t_ex parameter (tex t_ex = 1/(2kex2k_ex)) to the dispersion specific analysis. This is needed for the Ishima and Torchia 1999 2-site model for all timescales with pA p_A &#x226B; pBp_B. This commit follows step 5 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added support for the IT99 model to the relax_disp.cpmgfit_input user function. This is the Ishima and Torchia 1999 2-site model for all timescales with pA p_A &#x226B; pBp_B.

* Fix for the LM63 model for the relax_disp.cpmgfit_input user function. The grid search was incorrectly set up - the parameter is Tau, not text_ex.

* Fixed the scaling of the parameter text_ex.

* Basic fix for the checks of the Relax_disp.test_hansen_cpmgfit_input user function. The 'text_ex' parameter is now set as 'Tau'.

* Created the Relax_disp.test_hansen_cpmg_data_IT99 system test. This is for testing the Ishima and Torchia 1999 2-site model for all timescales with pA p_A &#x226B; pBp_B. This commit follows step 7 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).

* Added support for converting between kex k_ex and text_ex, and pA p_A and pB p_B for the dispersion analysis. This is performed by the new specific_analyses.relax_disp.parameters.param_conversion() function. For this, most of the code from the assemble_param_vector() function has been shifted into get_value(), and most of disassemble_param_vector() into set_value(). The dispersion analysis now also has a custom sim_init_values() method to handle these parameters.* Added support for calculating auxiliary parameter errors for the dispersion analysis. This is via the monte_carlo.error_analysis user function. The errors for the parameter pairs kexk_ex-tex t_ex and pAp_A-pB p_B for the non-model parameter are now calculated as well.* Fix for the dispersion auto-analysis - pA p_A and pB p_B parameters are no longer output for the IT99 model. These are not parameters of this model.

* Updated the lib.dispersion.ns_2site_star module with additional information from Paul Schanda. The details come from http://thread.gmane.org/gmane.science.nmr.relax.devel/4132/focus=4135. The exchange-free R2 R₂ value parameter names have been changed to match the convention of the other lib.dispersion modules.

* Added support for the R2B R_2B⁰ parameter as required by the 'NS 2-site star' model. This is the model of the numerical solution for the 2-site Bloch-McConnell equations using complex conjugate matrices. This commit follows step 5 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).

* Better support for the R2A R_2A⁰ and R2B R_2B⁰ relaxation rate parameters in the relaxation dispersion analysis. This includes a number of fixes to allow these two parameters to be handled correctly.* Added parameter conversions to go from pA p_A and kex k_ex to kge k_ge and kegk_eg. This is for the 'NS 2-site star' numerical model. The conversions have been added to the start of the target function to minimise mathematical operations to speed up the code.

* Improvement of the error handling in the 'NS 2-site star' model. The fA and pB p_B parameters are no longer being checked. Instead a Mgx value of 0.0 is being checked for. This catches additional problems. And now instead of the R2eff R_2eff value being set to zero, it is set to 1e99. This is because log of zero is -inf, and then multiplied by a negative constant gives positive inf.

* Added the 'NS 2-site star red' model to the specific_analyses.relax_disp.variables module. This is the model of the numerical solution for the 2-site Bloch-McConnell equations using complex conjugate matrices, whereby the simplification R20A R_2A⁰ = R20B R_2B⁰ is assumed. This commit follows step 1 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).

* Added the 'NS 2-site star red' model to the relax_disp.select_model user function frontend. This is the model of the numerical solution for the 2-site Bloch-McConnell equations using complex conjugate matrices, whereby the simplification R20A R_2A⁰ = R20B R_2B⁰ is assumed. This commit follows step 2 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).

* Created the 'NS 2-site star red' model target function. This is the model of the numerical solution for the 2-site Bloch-McConnell equations using complex conjugate matrices, whereby the simplification R20A R_2A⁰ = R20B R_2B⁰ is assumed. The code in common with the 'NS 2-site star' model has been shifted into the new calc_ns_2site_star_chi2() method. This commit follows step 4 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added support for the 'NS 2-site star red' model to the relax_disp.select_model user function back end. This is the model of the numerical solution for the 2-site Bloch-McConnell equations using complex conjugate matrices, whereby the simplification R20A R_2A⁰ = R20B R_2B⁰ is assumed. This commit follows step 6 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the 'NS 2-site star red' model to the relax user manual. This is the model of the numerical solution for the 2-site Bloch-McConnell equations using complex conjugate matrices, whereby the simplification R20A R_2A⁰ = R20B R_2B⁰ is assumed. This commit follows step 2 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).

* Added the 'CR72 red' model to the specific_analyses.relax_disp.variables module. This is the Carver and Richards 1972 analytic model with the simplification R20A R_2A⁰ = R20BR_2B⁰. The current 'CR72' makes the same assumption, but that model will be expanded to support R20A R_2A⁰ and R20B R_2B⁰ later. This commit follows step 1 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the 'CR72 red' model to the relax_disp.select_model user function frontend. This is the Carver and Richards 1972 analytic model with the simplification R20A R_2A⁰ = R20BR_2B⁰. The current 'CR72' makes the same assumption, but that model will be expanded to support R20A R_2A⁰ and R20B R_2B⁰ later. This commit follows step 2 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Created the 'CR72 red' model target function. This is the Carver and Richards 1972 analytic model with the simplification R20A R_2A⁰ = R20BR_2B⁰. The current 'CR72' makes the same assumption, but that model will be expanded to support R20A R_2A⁰ and R20B R_2B⁰ later. The code in common with the CR72 model has been shifted into the new calc_CR72_chi2() method. This commit follows step 4 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added support for the 'CR72 red' model to the relax_disp.select_model user function back end. This is the Carver and Richards 1972 analytic model with the simplification R20A R_2A⁰ = R20BR_2B⁰. The current 'CR72' makes the same assumption, but that model will be expanded to support R20A R_2A⁰ and R20B R_2B⁰ later. This commit follows step 6 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the 'CR72 red' model to the relax user manual. This is the Carver and Richards 1972 analytic model with the simplification R20A R_2A⁰ = R20BR_2B⁰. The current 'CR72' makes the same assumption, but that model will be expanded to support R20A R_2A⁰ and R20B R_2B⁰ later. This commit follows step 2 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).

* Expanded the CR72 model to handle both the R20A R_2A⁰ and R20B R_2B⁰ parameters. The 'CR72 red' model now functions as the old CR72 model where R20A R_2A⁰ = R20B R_2B⁰ = R20R₂⁰. All parts of the code have been modified. The lib.dispersion.cr72.r2eff_CR72() function has been expanded to support the full Carver and Richards 1972 equations, dropping back to the simplified form if R20A R_2A⁰ = R20BR_2B⁰.* Fix for the dispersion specific loop_parameters() method for the R20A R_2A⁰ and R20B R_2B⁰ parameters. The frequency index is now correctly returned for these and the R20 R₂⁰ parameter.* Better printouts of the R20A R_2A⁰ and R20B R_2B⁰ parameters at the end of minimisation.

* Small speed up for the lib.dispersion.cr72 module for the R20A R_2A⁰ != R20B R_2B⁰ case. Replicated calculations have been minimised.* Added support for model nesting in the relaxation dispersion auto-analysis. This involves copying the parameters from the simpler nested model rather than performing a full grid search. This is currently used to handle all models with R20A R_2A⁰ and R20B R_2B⁰ parameters where a simpler model with the single R20 R₂⁰ parameter is optimised first.

* Added the 'NS 2-site' R2eff R_2eff calculating function to the relax library. This is the model of the numerical solution for the 2-site Bloch-McConnell equations. It originates as optimization function number 1 from the fitting_main_kex.py script from Mathilde Lescanne, Paul Schanda, and Dominique Marion (see http://thread.gmane.org/gmane.science.nmr.relax.devel/4138, https://gna.org/task/?7712#comment2 and https://gna.org/support/download.php?file_id=18262). This commit follows step 3 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Updates and fixes for the lib.dispersion.ns_2site module. The function has been renamed, and the R1 R₁ arguments default to 0.0. The flip angle for the from the pulse.

* Modified the df, fA, and fB parameters to match the relax omega conventions of dw&delta;&omega;, wA, and wB. This follows from Paul Schanda's confirmation at http://thread.gmane.org/gmane.science.nmr.relax.devel/4132/focus=4159.* Speed up for the lib.dispersion.ns_matrices.rcpmg_3d() function. The pA p_A and pB p_B parameters are now sent into the function rather than being recreated by the function.* More changes to the numerical solution dispersion code to match relax's conventions. This includes the changes of df->dw&delta;&omega;, fA->wA, fB->wB, and Mgx->Mx.* Added the 'NS 2-site 3D red' model to the specific_analyses.relax_disp.variables module. This is the 'NS 2-site 3D' model with R20A R_2A⁰ = R20B R_2B⁰ = R20R₂⁰. This commit follows step 1 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the 'NS 2-site 3D red' model to the relax_disp.select_model user function frontend. This is the 'NS 2-site 3D' model with R20A R_2A⁰ = R20B R_2B⁰ = R20R₂⁰. This commit follows step 2 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Created the 'NS 2-site 3D red' model target function. This is the 'NS 2-site 3D' model with R20A R_2A⁰ = R20B R_2B⁰ = R20R₂⁰. This commit follows step 4 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added support for the 'NS 2-site 3D red' model to the relax_disp.select_model user function back end. This is the 'NS 2-site 3D' model with R20A R_2A⁰ = R20B R_2B⁰ = R20R₂⁰. This commit follows step 6 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).

* Added the 'NS 2-site expanded' R2eff R_2eff calculating function to the relax library. This is the numerical model for the 2-site Bloch-McConnell equations expanded using Maple by Nikolai Skrynnikov. It originates as optimization function number 5 from the fitting_main_kex.py script from Mathilde Lescanne, Paul Schanda, and Dominique Marion (see http://thread.gmane.org/gmane.science.nmr.relax.devel/4138, https://gna.org/task/?7712#comment2 and https://gna.org/support/download.php?file_id=18262). This commit follows step 3 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Fix for the lib.dispersion.ns_2site_expanded.r2eff_ns_2site_expanded() function. The pg variable should have been pA p_A and it needs to be sent into the function.

* Fix for the relax_disp.select_model user function for the 'NS 2-site expanded' model. There is only one R20 R₂⁰ parameter as R20A R_2A⁰ = R20B R_2B⁰ in this model.

* Large renaming of the relaxation dispersion models. This includes both the analytic and numerical models. All of the models with separate R20A R_2A⁰ and R20B R_2B⁰ parameters now have ' full' added to the end of the model name. And all of the corresponding reduced models whereby R20A R_2A⁰ = R20B R_2B⁰ = R20 R₂⁰ have had the trailing ' red' removed. All descriptions and variable names have been updated to match.

* Fixes for all of the definitions of the k_AB k_AB and k_BA k_BA exchange parameters. These were inverted in all parts of relax. The changes only affect the numerical dispersion models.

* Added an upper constraint of 2e6 rad/s for the kex k_ex dispersion parameter. This is to prevent slow optimisation to values in the order of 1e20 for models which fail.

* Fix for the Grace string for the dw &delta;&omega; dispersion parameter.* Updated the parameter value checks in the Relax_disp.test_hansen_cpmg_data_CR72 system test. The low precision parameters are slightly different because of the new upper constraint on kexk_ex, simply because optimisation is terminated early rather than optimisation giving different results.

* All of the numeric dispersion models are now much more robust. The real part of the magnetization vector for the A state could, for some parameter combinations, be either negative or NaN. These situations are now caught, and the R2eff R_2eff value set to a very large number.

* Added the 'LM63 3-site' R2eff R_2eff calculating function to the relax library. This is the Luz and Meiboom 1963 analytic model for three exchanging sites. This commit follows step 3 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).

* The dispersion auto-analysis now outputs files for the kBk_B, kC, phi_ex_B&phi;_ex,B, and phi_ex_C &phi;_ex,C parameters. This is in the write_results() method and is for creating text files and Grace plots for the 'LM 3-site' model.

* Added scripts for calculating the R2eff R_2eff values for the 'LM63 3-site' dispersion model test data.* Created the Relax_disp.test_r2eff_fit_fixed_time system test to show a failure in the auto-analysis. This shows a failure of the R2eff R_2eff fitting in the dispersion auto-analysis due to Monte Carlo simulations being run when the calc() function should be called.

* Made the 'LM63 3-site' dispersion model test data more realistic. Previously all the rates were within a few decimal places of the R20 R₂⁰ values. Now the dispersion is much more significant.

* Added the 'TP02' model to the specific_analyses.relax_disp.variables module. This is the Trott and Palmer 2002 R1rho R_1&rho; analytic model for 2-site exchange. This commit follows step 1 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the 'TP02' model to the relax_disp.select_model user function frontend. This is the Trott and Palmer 2002 R1rho R_1&rho; analytic model for 2-site exchange. This commit follows step 2 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the 'TP02' model to relaxation dispersion chapter of the relax user manual. This is the Trott and Palmer 2002 R1rho R_1&rho; analytic model for 2-site exchange. This commit follows step 2 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the 'TP02' R1rhoR_1&rho;' calculating function to the relax library. This is the Trott and Palmer 2002 R1rho R_1&rho; analytic model for 2-site exchange. This commit follows step 3 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907). The Matlab code from Skrynnikov and Tollinger has not been converted to Python code yet. This is to allow the Matlab->Python conversion to be followed.

* Created the 'TP02' model target function. This is the Trott and Palmer 2002 R1rho R_1&rho; analytic model for 2-site exchange. This commit follows step 4 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added support for the 'TP02' model to the relax_disp.select_model user function back end. This is the Trott and Palmer 2002 R1rho R_1&rho; analytic model for 2-site exchange. This commit follows step 6 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).

* Updated the M61 R1rho R_1&rho; model conditions in the table in the user manual.* Updated the TP02 R1rho R_1&rho; model conditions in the table in the manual. This cannot be fast exchange.

* Added a new user function to the Relax_disp.test_r1rho_off_res_fixed_time_tp02 system test. This is the chemical_shift.read user function which currently does not exist. Chemical shifts are needed to interpret off-resonance R1rho R_1&rho; data.

* Fix for converting dw &delta;&omega; from ppm to rad/s. Progress sr #3071: https://gna.org/support/?3071 - Implementation of Tollinger/Kay dispersion model (2001). Following the guide at: http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax.

* Added some more arguments to the dispersion target function class for off-resonance R1rho R_1&rho; models. This is the structure for the spin-lock offsets and the tilt angles for each spin.

* The TP02 model R1rho R_1&rho; off-resonance test data creation script now creates files of the R1 R₁ relaxation data. These files are needed for the system tests, as R1 R₁ data needs to be read.* The dispersion target function class now handles R1 R₁ relaxation data. This data is essential for the off-resonance R1rho R_1&rho; models.* The Relax_disp.test_r1rho_off_res_fixed_time_tp02 system test script now reads the R1 R₁ relaxation data. This data is essential for the model optimisation.* The dispersion specific code is now assembling the R1 R₁ data and passing it to the target function. The new specific_analyses.relax_disp.disp_data.return_r1_data() function creates a data structure holding all the R1 R₁ data. This is used in the off-resonance R1rho R_1&rho; experiments.

* Fix to allow R1 R₁ data to be randomised for Monte Carlo simulations for off-resonance R1rho R_1&rho; data. This is a temporary kludge for the dispersion analysis and needs to be replaced by a cleaner solution via the base_data_loop() method.

* Many fixes and improvements for all of the R1rho R_1&rho; dispersion models in the user manual. The equations are now correct and the parameter table updated with new parameters and equations.* Removed the unused theta and R1 R₁ arguments for the lib.dispersion.m61.r1rho_M61() function. These off-resonance parameters are not used in the on-resonance model.

* Added an R1 R₁ data file to r1rho_on_res_m61, by copying from the r1rho_off_res_tp02 test suite data.* Updated the Relax_disp.test_r1rho_on_res_fixed_time_dpl94 system test for off-resonance data. The offsets, R1 R₁ data, and chemical shifts are now setup or read by the script.* Fixes for the DPL94 model to make it truly off-resonance. The tilt angles and R1 R₁ data are now used by the target function.

* Renamed all of the current numeric dispersion models in relax to be specific to CPMG-type data. This is in preparation for adding R1rho R_1&rho; numeric models. It was proposed at http://thread.gmane.org/gmane.science.nmr.relax.devel/4461.* Added the 'NS R1rho 2-site' model to the specific_analyses.relax_disp.variables module. This is the numerical model for the 2-site Bloch-McConnell equations for R1rho R_1&rho; data. This commit follows step 1 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the 'NS R1rho 2-site' model to the relax_disp.select_model user function frontend. This is the numerical model for the 2-site Bloch-McConnell equations for R1rho R_1&rho; data. This commit follows step 2 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the 'NS R1rho 2-site' R1rho R_1&rho; calculating function to the relax library. This is the numerical model for the 2-site Bloch-McConnell equations for R1rho R_1&rho; data. This code originates from the Skrynikov & Tollinger code (the sim_all.tar file https://gna.org/support/download.php?file_id=18404 attached to https://gna.org/task/?7712#comment5). Specifically the funNumrho.m file. This commit follows step 3 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).

* Created the 'NS R1rho 2-site' model target function. This is the numerical model for the 2-site Bloch-McConnell equations for R1rho R_1&rho; data. The code originates from the funNumrho.m file from the Skrynikov & Tollinger code (the sim_all.tar file https://gna.org/support/download.php?file_id=18404 attached to https://gna.org/task/?7712#comment5). This commit follows step 4 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added support for the 'NS R1rho 2-site' model to the relax_disp.select_model user function back end. This is the numerical model for the 2-site Bloch-McConnell equations for R1rho R_1&rho; data. The code originates from the funNumrho.m file from the Skrynikov & Tollinger code (the sim_all.tar file https://gna.org/support/download.php?file_id=18404 attached to https://gna.org/task/?7712#comment5). This commit follows step 6 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).* Added the 'NS R1rho 2-site' model to the relax user manual. This is the numerical model for the 2-site Bloch-McConnell equations for R1rho R_1&rho; data. The code originates from the funNumrho.m file from the Skrynikov & Tollinger code (the sim_all.tar file https://gna.org/support/download.php?file_id=18404 attached to https://gna.org/task/?7712#comment5). This commit follows step 2 of the relaxation dispersion model addition tutorial (http://thread.gmane.org/gmane.science.nmr.relax.devel/3907).

* Created the Relax_disp.test_r1rho_ns_r1rho_2site_to_tp02 system test for the new 'NS R1rho 2-site' model. This tests the 'NS R1rho 2-site' model against the R1rho R_1&rho; off-resonance test data from the 'TP02' model.* A number of fixes for the 'NS R1rho 2-site' dispersion model. The model should now be fully functional. The chemical shift and R1 R₁ related data are now assembled for this model, and the data correctly passed from the target function to the lib.dispersion module.

* Better support for off-resonance R1rho R_1&rho; data in the dispersion GUI. A new row of buttons has been added to the dispersion GUI, just above the Peak list GUI element. The first button is for the spin.isotope user function and replaces the old GUI element. Two new buttons for loading R1 R₁ data and chemical shifts have also been added, as required for off-resonance R1rho R_1&rho; data.

* Small changes to the tooltips of the R1 R₁ and chemical shift buttons.

* Added the 'TP02' and 'NS R1rho 2-site' models to the R1rho R_1&rho; model list in the dispersion GUI. These models were missing from the list.

* Added the k_AB k_AB and k_BA k_BA parameters to the table of all dispersion parameters.

* Added the relax_disp.spin_lock_offset user function to the peak intensity wizard of the GUI. This is only for R1rhoR_1&rho;-type data and allows off-resonance data to be analysed in the GUI.

* Created the Relax_disp.test_tp02_data_to_tp02 GUI test. This is based on the system test of the same name. This GUI test checks that an off-resonance R1rho R_1&rho; analysis is functional in the GUI.

* Fixes for the units in the dispersion parameter table in the user manual. The units for dw &delta;&omega; are rad.s^-1 when used in the equations, but it is stored internally as ppm.

* Added script files for generating a saved state file with R2eff R_2eff values. Progress sr #3071: https://gna.org/support/?3071 - Implementation of Tollinger/Kay dispersion model (2001). Following the guide at: http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax.

* Modified the dispersion GUI analysis to handle the relax_disp.exp_type user function changes. This follows from http://thread.gmane.org/gmane.science.nmr.relax.devel/4530, the thread about supporting multiple data types such as SQ+MQ data simultaneously. The experiment type GUI element has been removed, the CPMG and R1rho R_1&rho; model lists merged into one, and much code for the experiment type removed. The peak analysis wizard will need to be heavily modified for the changes.

* Fix for the Relax_disp.test_hansen_cpmgfit_input system test for a new set of errors. The Hansen R2eff R_2eff values have been recalculated and the errors are now slightly different.

* Redesigned the peak intensity loading GUI wizard for handling multiple experiment types. This follows from http://thread.gmane.org/gmane.science.nmr.relax.devel/4530, the thread about supporting multiple data types such as SQ+MQ data simultaneously. The logic for the page ordering needed to be changed to be more dynamic. The CPMG and R1rho R_1&rho; pages need to be shown only if the corresponding experiment type exists in the current data pipe. Hence the has_cpmg_exp_type() and has_r1rho_exp_type() dispersion functions are now used by the new methods wizard_page_after_relax_time() and wizard_page_after_cpmg_frq(). A number of now useless flags have also been removed.* Added some sanity checks to the dispersion target function class. R1rho R_1&rho; models cannot be used with CPMG-type experiments, and CPMG models cannot be used with R1rhoR_1&rho;-type experiments.

* The GUI now warns if the user choses inappropriate models. When clicking on 'Execute', an error message appears if R1rho R_1&rho; models are selected for CPMG data and vice versa. This is simply for more intuitive user feedback.

* Added two functions for determining if a spectrum ID corresponds to a CPMG or R1rho R_1&rho; experiment. These are functions in specific_analyses.relax_disp.disp_data and they are called is_cpmg_exp_type() and is_r1rho_exp_type().

* Created a relax save file with just R2eff R_2eff values for the r1rho_on_res_m61 dispersion system tests data.

* The R2eff R_2eff result save file for the r1rho_on_res_m61 dispersion data now contains the full data set. The previous file did not contain the full exponential curves.

* Fixes for some of the R1rho R_1&rho; dispersion system test scripts. The relaxation time must be set for the reference spectrum.

* Converted references of ka and kA to k_ABk_AB. Progress sr #3071: https://gna.org/support/?3071 - Implementation of Tollinger/Kay dispersion model (2001). Following the guide at: http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax.

* Added k_AB k_AB to parameters. Progress sr #3071: https://gna.org/support/?3071 - Implementation of Tollinger/Kay dispersion model (2001). Following the guide at: http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax.

* Updated the Grace string for the k_AB k_AB parameter - it was being shown as kA.

* Added the k_AB k_AB and k_BA k_BA conversion equations to the dispersion parameter table in the user manual.

* Speed ups for the optimisation of all of the R1rho R_1&rho; dispersion models. The spin-lock field strength data structure is now converted from Hz to rad.s^-1 in the dispersion target function initialisation. Previously the conversion was happening multiple times per target function call. This has a noticeable effect on the test suite timings.* Some small speed ups for the TP02 R1rho R_1&rho; dispersion model optimisation. Some unneeded calculations and aliases were removed.* Added the write-out of 'dw&delta;&omega;' and 'k_ABk_AB' for model TSMFK01, when performing auto-analysis. Progress sr #3071: https://gna.org/support/?3071 - Implementation of Tollinger/Kay dispersion model (2001). Following the guide at: http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax.

* Added the conversion to k_AB k_AB from kex k_ex and pAp_A. k_AB k_AB = kex k_ex * (1.0 - pAp_A). Progress sr #3071: https://gna.org/support/?3071 - Implementation of Tollinger/Kay dispersion model (2001). Following the guide at: http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax.* Some more speed ups for the R1rho R_1&rho; dispersion models. For many models, the square of the spin-lock field strength is a part of the equations. Therefore this is now pre-calculated when the target function is initialised.

* Fix for the dispersion model list in the GUI - the R1rho R_1&rho; models were mixed up.* Added a sample script for an off-resonance R1rho R_1&rho; dispersion analysis.

* Fixed bug, where kex k_ex to k_AB k_AB where not possible if the model does not contain parameter 'pA'p_A. The conversion is now skipped. Progress sr #3071: https://gna.org/support/?3071 - Implementation of Tollinger/Kay dispersion model (2001). Following the guide at: http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax.* Added the conversion to k_BA k_BA from kex k_ex and pAp_A. Progress sr #3071: https://gna.org/support/?3071 - Implementation of Tollinger/Kay dispersion model (2001). Following the guide at: http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax.* Added system test for testing conversion to k_BA k_BA from kex k_ex and pAp_A. Progress sr #3071: https://gna.org/support/?3071 - Implementation of Tollinger/Kay dispersion model (2001). Following the guide at: http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax.

* Added system tests for conversion of kex k_ex to k_ABk<sub>AB</k_BA sub>/k_BA for models where kex k_ex and pA p_A is present. Progress sr #3071: https://gna.org/support/?3071 - Implementation of Tollinger/Kay dispersion model (2001). Following the guide at: http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax.

* The relax_disp.catia_input user function now creates the main CATIA input file and all R2eff R_2eff data files.

* The main CATIA input file requires the chemical shifts and R1 R₁ values to be fixed, even when missing. This is for the relax_disp.catia_input user function.

* Added the 'MQ NS CPMG 2-site' model to the relax_disp.select_model user function frontend. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_relax_disp.select_model_user_function_front_end Two new sections were added to the user function docstring for the MQ CPMG and MQ R1rho R_1&rho; experiment types.

* Added support for the new 'dwH&delta;&omega;_H' dispersion parameter. This is needed for the 'MQ NS CPMG 2-site' model support. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#Adding_support_for_the_parameters.

* Added the multi-quantum CPMG and R1rho R_1&rho; experiment types to the dispersion variables. This is needed for the 'MQ NS CPMG 2-site' model. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#Creating_a_new_experiment_type.* Added relaxation dispersion model lists for the MQ R1rho R_1&rho; models. These are stubs as no MQ R1rho R_1&rho; models are yet supported by relax.

* Added the 'MQ NS CPMG 2-site' R1rho R_1&rho; calculating function to the relax library. This is the 2-site numeric solution for multi-quantum CPMG-type data. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_relax_library.

* Added a latex definition for the dwH &delta;&omega;_H dispersion parameter and added the 'MQ' abbreviation.

* Added some value.set calls to the Relax_disp.test_sprangers_cpmg_data_auto_analysis system test. These user function calls will be used to test a new concept of fixing parameters in the grid search. The dw &delta;&omega; and dwH &delta;&omega;_H parameters are fixed to the experimental values, as described in the README file from Remco Sprangers (in test_suite/shared_data/dispersion/Sprangers_ClpP).

* A bit of help for some of the R1rho R_1&rho; dispersion model system tests. These now fail after a fundamental fix. The problem is only due to the very coarse grid search size - a finer grid search allows the solution to be correctly found. However as this is far too slow, instead the kex k_ex parameter is set to be close to the solution to skip a grid search dimension.

* Simplified the Relax_disp.test_sprangers_cpmg_data_auto_analysis system test script. The pA p_A and kex k_ex parameters are now also pre-set to speed things up.

* Fixes for the R2eff R_2eff data files for Sprangers ClpP data.* Artificially increased the errors in Sprangers ClpP data to match the publication. The R2eff R_2eff errors are simply multiplied by 5, as the errors from the paper cannot be replicated.

* Added the 'MQ CR72' R2eff R_2eff calculating function to the relax library. This is the Carver and Richards (1972) 2-site model expanded for MQ CPMG data by Korzhnev et al., 2004. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_relax_library. The corresponding target function was updated to input the correct arguments.

* Bug fix for the dispersion specific loop_parameters() function for the multiple quantum models. The dw &delta;&omega; and dwH &delta;&omega;_H parameters were being interleaved rather than all dw &delta;&omega; for all spins first and then all dwH&delta;&omega;_H. The result was that these parameters were being mixed up in the MQ model target functions when clustering was activated, causing total failure of the MQ models.

* Modified the dwH &delta;&omega;_H symbol in the relax user manual.

* A large number of fixes for the relaxation dispersion system tested needed for the fix which changed the format of the keys by which the R2effR<sub>2eff</R1rho sub>/R_1&rho; data is accessed.

* Fixes for the calc user function for the dispersion analysis. This now does something logical for the non-R2eff models. The chi-squared value is now being calculated and stored. Previously this was only calculating the R2effR_2eff/R<sub>1&rho;</R1rho sub> values for fixed relaxation time period data for the 'R2eff' model and failing for all others. Now the pre-existing _back_calc_r2eff() method is used to back-calculate and store the chi-squared value.* Redesigned the Relax_disp.test_hansen_cpmg_data_to_ns_cpmg_2site_star_full system test. The R2A R_2A⁰ and R2B R_2B⁰ rates cannot be distinguished for this data, therefore there was no unique solution. This resulted in too much variability between 32 and 64-bit systems as well as different operating systems. Instead a single calc user function call is used to determine the chi-squared value for a fixed set of parameters.

* Added test data where both the spin-lock time, the spin lock offset and the spin lock field is varied. The data is published in "Kjaergaard, M., Andersen, L., Nielsen, L.D. & Teilum, K. (2013). A Folded Excited State of Ligand-Free Nuclear Coactivator Binding Domain (NCBD) Underlies Plasticity in Ligand Recognition. Biochemistry, 52, 1686-1693" with experimental conditions that "off-resonance R1rho R_1&rho; relaxation dispersion experiments on 15N were recorded at 18.8 T and 31 C." and "using the pulse sequence of Mulder et al. with spin-lock field strengths from 431 to 1649 Hz and offsets ranging from 0 to 10000 Hz."

* Modified settings script for R1rho R_1&rho; test dataset.

* Correcting the R1rho R_1&rho; settings script for the right calculation of the spin-lock offset, omega_rf, in ppm when offset values are provided in Hz.

* Created a relax state for the R2eff R_2eff SQ data of Korzhnev et al., 2005.

* Created two new dispersion variables - EXP_TYPE_LIST_CPMG and EXP_TYPE_LIST_R1RHO. This will be used to simplify identifying CPMG vs. R1rho R_1&rho; data types.

* Turned off the dw&delta;&omega;, dwH &delta;&omega;_H > 0 constraint for the 'MMQ 2-site' model.

* Changed the definition of tex t_ex thanks to feedback from Nikolai Skrynnikov. This was previously defined as tex t_ex = 1/(2kex2k_ex) to be compatible with CPMGFit, but has now been changed to tex t_ex = 1/kexk_ex.* Converted the 'IT99' dispersion model parameters to pA p_A and dw&delta;&omega;. This is thanks to feedback from Nikolai Skrynnikov. I have no idea why the phi_ex &phi;_ex and pAp_A.dw&delta;&omega;^2 parameters were being used in the first place. The model results after the change are identical.

* A number of fixes for the script UI section of the dispersion chapter of the manual. The NUMERIC_ONLY variable is now explained and the R1rho R_1&rho; MODEL list has been changed to a set of reasonable models.

* Created the Relax_disp.test_hansen_cpmg_data_auto_analysis_r2eff system test. This is to test the full dispersion auto-analysis on Flemming Hansen's CPMG data using the original R2eff R_2eff data rather than the derived peak heights.

* Created files of R2eff R_2eff values and errors for Flemming Hansen's CPMG data.

* The error analysis is now skipped in the dispersion auto-analysis if the 'R2eff' model is not given. It is then assumed that R2effR<sub>2eff</R1rho sub>/R_1&rho; data has already been loaded into the base data pipe and hence the error analysis is not needed. This avoids fatal errors.

* Fix for the dispersion auto-analysis if R2eff R_2eff data already exists. The data is no longer copied from the non-existent 'R2eff' data pipe.* Fixes for the dispersion specific overfit_deselect() method for when R2eff R_2eff data is read. This now no longer checks for intensity data but rather R2eff R_2eff data, as intensity data will not be present if R2eff R_2eff data is directly read rather than peak intensities.* Fixes for the Relax_disp.test_hansen_cpmg_data_auto_analysis system test. The setup of the auto-analysis could be simplified as the base data pipe can now contain R2eff R_2eff data. The R2eff R_2eff data in the 'R2eff' data pipe was no longer being read.* Some small fixes to allow the optimisation of dispersion models when no peak intensity data has been read. This is for when R2eff R_2eff data has been read instead.* The relax_disp.insignificance user function can now handle selected spins with no R2effR_2eff/R<sub>1&rho;</R1rho sub> data.* Fixes for the Monte Carlo simulations in the dispersion analysis when R2eff R_2eff data has been read. As peak intensity data has not been read, the relaxation time period will not have been set. The _back_calc_r2eff() method can now handle this.* Improved the R2eff R_2eff errors for Flemming Hansen's CPMG data. The errors are now calculated using the data from all spins rather than a truncated subset. The errors will therefore be much more accurate.* Fix for return_index_from_disp_point() for when R2effR_2eff/R<sub>1&rho;</R1rho sub> data is loaded rather than intensities. This specific_analyses.relax_disp.disp_data.return_index_from_disp_point() function was always subtracting 1 from the dispersion point index to take the reference spectrum into account. This however fails if R2effR_2eff/R<sub>1&rho;</R1rho sub> data is loaded instead.* Fixes for the Relax_disp.test_hansen_cpmg_data_auto_analysis* system tests. The Relax_disp.test_hansen_cpmg_data_auto_analysis system test needed updating due to the more accurate R2eff R_2eff errors. The Relax_disp.test_hansen_cpmg_data_auto_analysis_r2eff system test also needed this change. It also no longer has a spin system for residue 4.

* Implemented the relax_disp.r2eff_read_spin user function. This allows R2effR<sub>2eff</R1rho sub>/R_1&rho; files for each spin to be read.

* Redesigned the relax_disp.r2eff_read and relax_disp.r2eff_read_spin user functions. These now no longer set the metadata (spectrometer frequency and experiment type) themselves. Instead an experiment ID string must be supplied. The spectrometer.frequency and relax_disp.exp_type user functions will therefore need to be called before these R2eff R_2eff functions.

* Conversion of the format of the relaxation dispersion R2effR<sub>2eff</R1rho sub>/R_1&rho; data structures. These are now lists of lists of lists of numpy arrays instead of pure numpy rank-4 arrays. This only affects a number of related data structures in the dispersion target function class. The main purpose is to prepare to have a different number of dispersion points per experiment, per spin, and per spectrometer frequency.

* Added more R1rho R_1&rho; model references to the bibliography file for the manual. This includes the Trott and Palmer 2004 N-site and the Miloushev and Palmer 2005 2-site models. The Trott and Palmer 2002 R1rho R_1&rho; model reference has been expanded to include all details.* Added the TP04 and MP05 R1rho R_1&rho; dispersion models to the manual. These are not implemented in relax, or any of the software in the software comparison section, but are included for completeness. This was pointed out by Art Palmer.

* Added the 'MP05' model to the dispersion variables. This is the Miloushev and Palmer 2005 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#Adding_the_model_to_the_list.* Added the 'MP05' model to the relax_disp.select_model user function frontend. This is the Miloushev and Palmer 2005 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_relax_disp.select_model_user_function_front_end.* Added support for the 'MP05' model to the relax_disp.select_model user function back end. This is the Miloushev and Palmer 2005 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_relax_disp.select_model_user_function_back_end.* Created the 'MP05' model target function. This is the Miloushev and Palmer 2005 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_target_function.* Added the 'MP05' R2eff R_2eff calculating function to the relax library. This is the Miloushev and Palmer 2005 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_relax_library. Just in case git-svn does not preserve the file copying history, the lib/dispersion/mp05.py file was copied from the tp02.py file.* Debugging of the 'MP05' dispersion model - optimisation is now setup correctly. This is the Miloushev and Palmer 2005 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax.html#Debugging.

* Added the 'MP05' model to the GUI model list. This is the Miloushev and Palmer 2005 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_GUI.* Added the 'MP05' model to the dispersion auto-analysis. This is the Miloushev and Palmer 2005 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_auto-analysis.* Added the 'MP05' model to the relax user manual. This is the Miloushev and Palmer 2005 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_relax_manual. The 'MP05' model was already partly in the manual, however it was listed as unimplemented. All of the tables and the dispersion chapter text has been updated for the model.* Modified the R1rho_analysis.py sample script to use the 'MP05' model. This is the Miloushev and Palmer 2005 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_sample_scripts.

* Removed all remnants of the MQ R1rho R_1&rho; data type. This data type does not exist and was mostly removed, but some small bits remained.* Created the Relax_disp.test_tp02_data_to_tap03 system test. This is the Trott et al, 2003 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_test_suite This was copied from the Relax_disp.test_tp02_data_to_mp05 system test.* Added the 'TAP03' model to the dispersion variables. This is the Trott, Abergel and Palmer 2003 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#Adding_the_model_to_the_list.* Added the 'TAP03' model to the relax_disp.select_model user function frontend. This is the Trott, Abergel and Palmer 2003 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_relax_disp.select_model_user_function_front_end.* Added support for the 'TAP03' model to the relax_disp.select_model user function back end. This is the Trott, Abergel and Palmer 2003 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_relax_disp.select_model_user_function_back_end.* Created the 'TAP03' model target function. This is the Trott, Abergel and Palmer 2003 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_target_function.* Added the 'TAP03' R2eff R_2eff calculating function to the relax library. This is the Trott, Abergel and Palmer 2003 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_relax_library.* Debugging of the 'TAP03' dispersion model - optimisation is now setup correctly. This is the Trott, Abergel and Palmer 2003 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax.html#Debugging.

* Added the 'TAP03' model to the GUI model list. This is the Trott, Abergel and Palmer 2003 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_GUI.* Added the 'TAP03' model to the dispersion auto-analysis. This is the Trott, Abergel and Palmer 2003 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_auto-analysis.* Added the 'TAP03' model to the relax user manual. This is the Trott, Abergel and Palmer 2003 R1rho R_1&rho; analytic model for 2-site off-resonance exchange. This follows the tutorial for adding relaxation dispersion models at http://wiki.nmr-relax.com/Tutorial_for_adding_relaxation_dispersion_models_to_relax#The_relax_manual. The 'TAP03' model was already partly in the manual, however it was listed as unimplemented. All of the tables and the dispersion chapter text has been updated for the model.

* Improvements to all of the R1rho R_1&rho; model descriptions in the dispersion chapter of the manual.

* Implemented the 'MMQ 2-site' CPMG model equations from the Korzhnev et al., 2005 reference. The paper reference is "Dmitry M. Korzhnev, Philipp Neudecker, Anthony Mittermaier, Vladislav Yu. Orekhov, and Lewis E. Kay (2005). Multiple-site exchange in proteins studied with a suite of six NMR relaxation dispersion experiments: An application to the folding of a Fyn SH3 domain mutant. J. Am. Chem. Soc., 127, 15602-15611. (doi: http://dx.doi.org/10.1021/ja054550e)". The original code from Mathilde Lescanne and Dominique Marion has only slightly been modified for this change as the MQ data treatment in the Korzhnev et al., 2004 reference is the same as in the 2005 reference, but using a different notation. This has been renamed to r2eff_mmq_2site_mq(). The new r2eff_mmq_2site_sq_dq_zq() function has been added to the lib.dispersion.mmq_2site module to allows the SQ, DQ, and ZQ R2eff R_2eff data to be calculated. This function follows the notation of the 2005 paper. The populate_matrix() function has been modified to only accept one combined chemical shift difference value. It can now also accept different values for R20A R_2A⁰ and R20BR_2B⁰, though the mmq_2site module defaults to R20AR_2A⁰=R20BR_2B⁰.* The r2eff_mmq_*() functions of lib.dispersion.mmq_2site now accept different R20A R_2A⁰ and R20B R_2B⁰ arguments. These are set to the same thing within the dispersion target function.* Converted the spin specific 'r2', 'r2a', and 'r2b' dispersion parameters from lists to dictionaries. The new parameter keys are based on the experiment type and the spectrometer frequency. These keys are supported by the generate_r20_key() and decompose_r20_key() pair of functions in the specific_analyses.relax_disp.disp_data module. This enables support for different R20 R₂⁰ parameters for each experiment type - a key piece of infrastructure for the MMQ models. The relax_disp.select_model user function backend was modified so the parameter list only contains one instance for each of the 'r2', 'r2a', or 'r2b' strings. The specific_analyses.relax_disp.parameters.loop_parameters() function was modified so that the R20key R₂⁰key rather than frequency index is returned for the R20 R₂⁰ parameters. Many other code changes were required.* The R20 R₂⁰ values are now correctly handled in the dispersion target function for MMQ-type data.

* The cpmg_fit script for the Korzhnev et al., 2005 15N ZQ CPMG data now starts at the relax solution. This is to try to find better solutions for dw &delta;&omega; and dwH&delta;&omega;_H, thought it was not so successful.

* Fixes for the cpmg_fit results for all of the data from Korzhnev et al., 2005. The dwH &delta;&omega;_H value must start negative, otherwise optimisation will fail to find the correct minimum.

* The 'MMQ 2-site' dispersion model can now be optimised if no heteronuclear R2eff R_2eff data is loaded.

* The R2eff R_2eff data key has been changed in the dispersion analysis. The experiment type has been added to the key so that R2eff R_2eff data is not mixed up when data from multiple experiments is present.

* Created the relax_disp.write_disp_curves user function. This is based on feedback from Nikolai Skrynnikov. The user function will generate one file per spin system and dump all of the R2eff R_2eff values (measured, back calculated, and errors) into the file.

* Better support for the MMQ-type data dispersion models for the end of the optimisation. The back calculated R2eff R_2eff values are now handled correctly for the attached proton in the spin system.

* The relaxation dispersion calculate user function now stores the back calculated R2eff R_2eff values. A number of changes were required for this. The code from the end of the Disp_result_command.run() method was converted to the function specific_analyses.relax_disp.disp_data.pack_back_calc_r2eff(). This allows the back calculation R2eff R_2eff unpacking code to be shared. The new has_proton_mmq_cpmg() function has also been created to simplify the code.

* Fix for the 'MQ CR72' model for MQ-type data. The check to prevent acos of a number less than 1 has been changed to switch the sign rather than to set the back calculated R2eff R_2eff to 1e99.* Another bug fix for the 'MQ CR72' dispersion model. The nu_CPMG value rather than the relaxation time was being used to calculate the R2eff R_2eff values as the division by 'n' was missing.

* Fix for the 'MQ CR72' dispersion model target function. The correct R20 R₂⁰ values are now extracted from the parameter vector.* Improvements for the 'CR72' and 'MQ CR72' dispersion model R2eff R_2eff calculating functions. The numpy.arccosh() function can handle all input values when complex, therefore the checks for the real part being above 1 are not necessary.

* Created a relax script to compare the 'MQ CR72' dispersion model results to cpmg_fit. The cpmg_fit solution is used as the input parameters for relax, and then a calc user function call is used to back calculate the R2eff R_2eff values. These values are then plotted to show the perfect match.

* Fix for the Relax_disp.test_hansen_cpmg_data_auto_analysis system test. The kex k_ex value check needed to be scaled back.

* Implemented model elimination for the relaxation dispersion analysis. This currently uses the pA p_A limits of 0.501 < pA p_A < 0.999 to determine if a model has failed. To implement this, the dispersion API methods deselect(), eliminate(), get_param_names() and get_param_values() were written. These were copied from the model-free analysis and modified as needed.

* The relaxation dispersion target function class can now handle cpmg_frqs arguments of None. This is useful for R1rho R_1&rho; models.

* Added the tex t_ex > 1.0 model elimination rule for the dispersion analysis.

* Fix for the specific_analyses.relax_disp.optimisation.back_calc_r2eff() function. The R2eff R_2eff error data structure when the cpmg_frqs or spin_lock_nu1 argument is supplied was all zeros, whereas it should all be ones. This was causing many divide by zero numpy warnings to appear on certain operating systems (Mac OS X).