updated data overview

data_overview_mod.py

@@ -69,7 +69,8 @@ def data_overview3():
 
     #####################################################
     #grid_lower_lower = gridspec.GridSpecFromSubplotSpec(1, 2, grid0[1], wspace = 0.5, hspace=0.55)#, height_ratios = [1,3]
-    cell_types = [' Ampullary', ' P-unit']#, ' P-unit',]#' P-unit',
+    cell_types = [' P-unit',' Ampullary', ]#, ' P-unit',]#' P-unit',
+    cell_types_name = ['P-units','Ampullary cells',]
     species = ' Apteronotus leptorhynchus'
     burst_corr_reset = 'response_modulation'
     burst_fraction = [1000, 1000]# 50,  # ,1,1]
@@ -86,7 +87,7 @@ def data_overview3():
               ]  # + '_diagonal_proj'
 
 
-    cell_types_name = ['Ampullary cells','P-units']
+
 
     for c, cell_type_here in enumerate(cell_types):
         frame_file = setting_overview_score(frame_load_sp, cell_type_here, min_amp='range', species=species)
@@ -143,9 +144,14 @@ def data_overview3():
                 ymin = 0
             xmin = 0
 
+            if (' P-unit' in cell_type_here) & (x_axis[v] == 'cv_base' ):
+                xlimk = [0, 1.7]
+            else:
+                xlimk = None
+
             cmap, _, y_axis = plt_burst_modulation_hists(axk, axl, var_item_names[v], axs, cell_type_here,
                                                          x_axis[v], frame_file, max_val, scores_here[v],
-                                                         burst_fraction=burst_fraction[c], ha = 'right', x_pos = 1, xmin = xmin, ymin = ymin, burst_fraction_reset = burst_corr_reset,var_item=var_type)
+                                                         burst_fraction=burst_fraction[c],xlim = xlimk, ha = 'right', x_pos = 1, xmin = xmin, ymin = ymin, burst_fraction_reset = burst_corr_reset,var_item=var_type)
 
             if v == 0:
                 colors = colors_overview()
@@ -155,8 +161,10 @@ def data_overview3():
             axl.show_spines('')
             axs.set_ylabel(score_name[v])
             axs.set_xlabel(x_axis_names[v])
-            if  (cell_type_here == ' P-unit') & (x_axis[v] == 'cv_base' ):
+            if  (' P-unit' in cell_type_here) & (x_axis[v] == 'cv_base' ):
-                axs.set_xlim(0,1.7)
+                axs.set_xlim(xlimk)
+                axk.set_xlim(xlimk)
+                #embed()
             #remove_yticks(axl)
 
             if log:

flowchart.py

@@ -5,7 +5,7 @@ import numpy as np
 import pandas as pd
 from plotstyle import plot_style, spines_params
 from utils_suseptibility import model_sheme_in_one
-from utils_all import default_figsize, model_sheme_split, remove_yticks, save_visualization
+from utils_all import default_figsize, model_sheme_split2, remove_yticks, save_visualization
 #from utils_all import default_settings, resave_small_files
 #from plt_RAM import model_and_data, model_and_data_sheme, model_and_data_vertical2
 import matplotlib.gridspec as gridspec
@@ -18,7 +18,7 @@ def flowchart():
     grid_orig = gridspec.GridSpec(1, 1, wspace=0.15, bottom=0.07,
                                   hspace=0.1, left=0.05, right=0.96,
                                   top=0.9)  # , height_ratios = [0.4,3]
-    model_sheme_split(grid_orig[0])  # grid_sheme grid_lower[3]
+    model_sheme_split2(grid_orig[0])  # grid_sheme grid_lower[3]
     # model_sheme_only(grid[2])
     save_visualization()
     #fig.savefig()

model_and_data.py

@@ -47,8 +47,8 @@ def model_and_data2(eod_metrice = False, width=0.005, nffts=['whole'], powers=[1
     cells_all = [good_data[0]]
 
     plot_style()
-    default_figsize(column=2, length=4.75)  # 0.75
+    default_figsize(column=2, length=3.25)  #4.75 0.75
-    grid = gridspec.GridSpec(3, 4, wspace=0.85, bottom=0.07,
+    grid = gridspec.GridSpec(2, 4, wspace=0.85, bottom=0.07,
                              hspace=0.18, left=0.09, right=0.93, top=0.94)
 
     a = 0
@@ -85,7 +85,7 @@ def model_and_data2(eod_metrice = False, width=0.005, nffts=['whole'], powers=[1
         #################################
         # data cells
         # embed()
-        grid_data = gridspec.GridSpecFromSubplotSpec(1, 1, grid[0, 1],
+        grid_data = gridspec.GridSpecFromSubplotSpec(1, 1, grid[0, 0],
                                                      hspace=hs)
 
         #ypos_x_modelanddata()
@@ -129,12 +129,12 @@ def model_and_data2(eod_metrice = False, width=0.005, nffts=['whole'], powers=[1
             'calc_RAM_model-2__nfft_whole_power_1_RAM_additiv_cv_adapt_factor_scaled_cNoise_0.1_cSig_0.9_cutoff1_300_cutoff2_300no_sinz_length1_TrialsStim_11_a_fr_1__trans1s__TrialsNr_1_fft_o_forward_fft_i_forward_Hz_mV',
             'calc_RAM_model-2__nfft_whole_power_1_RAM_additiv_cv_adapt_factor_scaled_cNoise_0.1_cSig_0.9_cutoff1_300_cutoff2_300no_sinz_length1_TrialsStim_500000_a_fr_1__trans1s__TrialsNr_1_fft_o_forward_fft_i_forward_Hz_mV',
 
-            'calc_RAM_model-2__nfft_whole_power_1_afe_0.009_RAM_RAM_additiv_cv_adapt_factor_scaled_cNoise_0.1_cSig_0.9_cutoff1_300_cutoff2_300no_sinz_length1_TrialsStim_11_a_fr_1__trans1s__TrialsNr_1_fft_o_forward_fft_i_forward_Hz_mV',
-            'calc_RAM_model-2__nfft_whole_power_1_afe_0.009_RAM_RAM_additiv_cv_adapt_factor_scaled_cNoise_0.1_cSig_0.9_cutoff1_300_cutoff2_300no_sinz_length1_TrialsStim_500000_a_fr_1__trans1s__TrialsNr_1_fft_o_forward_fft_i_forward_Hz_mV',
 
         ]
+        #'calc_RAM_model-2__nfft_whole_power_1_afe_0.009_RAM_RAM_additiv_cv_adapt_factor_scaled_cNoise_0.1_cSig_0.9_cutoff1_300_cutoff2_300no_sinz_length1_TrialsStim_11_a_fr_1__trans1s__TrialsNr_1_fft_o_forward_fft_i_forward_Hz_mV',
+        #'calc_RAM_model-2__nfft_whole_power_1_afe_0.009_RAM_RAM_additiv_cv_adapt_factor_scaled_cNoise_0.1_cSig_0.9_cutoff1_300_cutoff2_300no_sinz_length1_TrialsStim_500000_a_fr_1__trans1s__TrialsNr_1_fft_o_forward_fft_i_forward_Hz_mV',
 
-        nrs_s = [2, 3, 6, 7, 10, 11]
+        nrs_s = [2, 3, 6, 7]#, 10, 11
         #embed()
         tr_name = trial_nr/1000000
         if tr_name == 1:
@@ -142,8 +142,9 @@ def model_and_data2(eod_metrice = False, width=0.005, nffts=['whole'], powers=[1
         titles = ['Model\n$N=11$ \n $c=1\,\%$', 'Model\n$N=%s $' % (tr_name) +'\,million \n $c=1\,\%$',
                   'Model\,('+noise_name().lower()+')' + '\n' + '$N=11$\n $c=0\,\%$',
                   'Model\,('+noise_name().lower()+')' + '\n' + '$N=%s$' % (tr_name) + '\,million \n $c=0\,\%$',
-                  'Model\,('+noise_name().lower()+')' + '\n' + '$N=11$\n $c=1\,\%$',
+                  ]#%
-                  'Model\,('+noise_name().lower()+')' + '\n' + '$N=%s$' % (tr_name) + '\,million\n $c=1\,\%$ ']#%
+        #'Model\,('+noise_name().lower()+')' + '\n' + '$N=11$\n $c=1\,\%$',
+        #          'Model\,('+noise_name().lower()+')' + '\n' + '$N=%s$' % (tr_name) + '\,million\n $c=1\,\%$ '
         ax_model = []
 
         for s, sav_name in enumerate(save_names):
@@ -205,15 +206,15 @@ def model_and_data2(eod_metrice = False, width=0.005, nffts=['whole'], powers=[1
         #################################################
         # Flowcharts
 
-        var_types = ['', 'additiv_cv_adapt_factor_scaled', 'additiv_cv_adapt_factor_scaled']
+        var_types = ['',  'additiv_cv_adapt_factor_scaled']#'additiv_cv_adapt_factor_scaled',
         ##additiv_cv_adapt_factor_scaled
-        a_fes = [0.009, 0, 0.009]
+        a_fes = [0.009, 0]#, 0.009
-        eod_fe = [750, 750, 750]
+        eod_fe = [750, 750]#, 750
         ylim = [-0.5, 0.5]
-        c_sigs = [0, 0.9, 0.9]
+        c_sigs = [0, 0.9]#, 0.9
-        grid_left = [[], grid[1, 0], grid[2, 0]]
+        grid_left = [[], grid[1, 0]]#, grid[2, 0]
         ax_ams = []
-        for g, grid_here in enumerate([grid[0, 0], grid[1, 0], grid[2, 0]]):
+        for g, grid_here in enumerate([grid[0, 1], grid[1, 1]]):#, grid[2, 0]
             grid_lowpass = gridspec.GridSpecFromSubplotSpec(4, 1,
                                                             subplot_spec=grid_here, hspace=0.3,
                                                             height_ratios=[1, 1,1, 0.1])
@@ -229,7 +230,7 @@ def model_and_data2(eod_metrice = False, width=0.005, nffts=['whole'], powers=[1
             noise_final_c, spike_times, stimulus, stimulus_here, time, v_dent_output, v_mem_output, frame = get_flowchart_params(
                 a_fes, a_fr, g, c_sigs[g], cell, deltat, eod_fr, model_params, stimulus_length, v_offset, var_types,
                 eod_fe=eod_fe)
-            # embed()
+
 
             if (len(np.unique(frame.RAM_afe)) > 1) & (len(np.unique(frame.RAM_noise)) > 1):
                 grid_lowpass2 = gridspec.GridSpecFromSubplotSpec(4, 1,
@@ -307,7 +308,7 @@ def model_and_data2(eod_metrice = False, width=0.005, nffts=['whole'], powers=[1
             ax_n, ff, pp, ff_am, pp_am = plot_lowpass2([grid_lowpass[2]], time, noise_final_c, deltat, eod_fr,
                                                        extract=False, color1='grey', lw=1)
             remove_yticks(ax_n)
-            if g == 2:
+            if g == 1:
                 ax_n.set_xlabel('Time [ms]', labelpad = -0.5)
             else:
                 remove_xticks(ax_n)
@@ -330,25 +331,26 @@ def model_and_data2(eod_metrice = False, width=0.005, nffts=['whole'], powers=[1
                 ax_external.text(1, 1, 'RAM', ha='right', color='red', transform=ax_external.transAxes)
                 ax_intrinsic.text(start_pos_modeldata(), 1, signal_component_name(), ha='right', color='purple',
                                   transform=ax_intrinsic.transAxes)
-
+            #embed()
         set_same_ylim(ax_ams, up='up')
         # embed()
         axes = np.concatenate([ax_data, ax_model])
 
-        axes = [ax_ams[0], axes[0], axes[1], axes[2], ax_ams[1], axes[3], axes[4], ax_ams[2], axes[5],
+        axes = [ax_ams[0], axes[1], axes[2], ax_ams[1], axes[3], axes[4],  ]#ax_ams[2], axes[5], axes[6],
-                axes[6], ]
+        #axd1 = plt.subplot(grid[1, 1])
-        axd1 = plt.subplot(grid[1, 1])
+        #axd2 = plt.subplot(grid[2, 1])
-        axd2 = plt.subplot(grid[2, 1])
         #ax_data.extend([,])
-        axd1.show_spines('')
+        #axd1.show_spines('')
-        axd2.show_spines('')
+        #axd2.show_spines('')
         #embed()
         #axes = [[ax_ams[0],ax_data[0],axes[2], axes[3]],[ax_ams[1],axd1,axes[4], axes[5]],[axd2,axd2, axes[6],  axes[7]]]
-        fig.tag([axes[0:4]], xoffs=-3, yoffs=1.6)  # ax_ams[3],
+
-        fig.tag([[axes[4]]], xoffs=-3, yoffs=1.6, minor_index=0)  # ax_ams[3],
+        fig.tag([ax_data], xoffs=-3, yoffs=1.6)  # ax_ams[3],
-        fig.tag([axes[5:7]], xoffs=-3, yoffs=1.6, major_index = 1, minor_index = 2)  # ax_ams[3],
+        fig.tag([axes[0:3]], xoffs=-3, yoffs=1.6)  # ax_ams[3],
-        fig.tag([[axes[7]]], xoffs=-3, yoffs=1.6, major_index=2,minor_index=0)  # ax_ams[3],
+        #fig.tag([[axes[4]]], xoffs=-3, yoffs=1.6, minor_index=0)  # ax_ams[3],
-        fig.tag([axes[8::]], xoffs=-3, yoffs=1.6, major_index=2, minor_index=2)  # ax_ams[3],
+        fig.tag([axes[3:6]], xoffs=-3, yoffs=1.6)  #, major_index = 1, minor_index = 2 ax_ams[3],
+        #fig.tag([[axes[7]]], xoffs=-3, yoffs=1.6, major_index=2,minor_index=0)  # ax_ams[3],
+        #fig.tag([axes[8::]], xoffs=-3, yoffs=1.6, major_index=2, minor_index=2)  # ax_ams[3],
         #fig.tag([axes[7::]], xoffs=-3, yoffs=1.6)  # ax_ams[3],
 
         #fig.tag([ax_ams[0],ax_data[0],axes[2], axes[3]], xoffs=-3, yoffs=1.6)#ax_ams[3],

susceptibility1.tex.bak

@@ -623,8 +623,9 @@ The nonlinearity in this work were found in low-CV P-units. For this nonlinear e
 
 
 %\subsubsection{A heterogeneous readout is required to cover all female-intruder combinations} 
-A heterogeneous readout might be not only physiologically plausible but also required since nonlinear effects depending on cell property \fbase{} and not on stimulus properties might be not behaviorally relevant. Nonlinear effects might facilitate the encoding of faint signals during a three fish setting, the electrosensory cocktail party. The EOD frequencies of the encoutered three fish would be drawn from the EOD frequency distribution of these fish and a stable faint signal detection would require a response irrespective of the individual EOD frequencies. In this work, nonlinear effects were always found only for specific frequencies in relation to \fbase{}, corresponding to findings from previous literature \citealp{Voronenko2017}. Weather integrating from a heterogeneous population with different \fbasesolid{} (50--450\,Hz, \citealp{Grewe2017, Hladnik2023}) would cover the behaviorally relevant range in the electrosensory cocktail party should be addressed in further studies. 
+A heterogeneous readout might be not only physiologically plausible but also required since nonlinear effects depending on cell property \fbase{} and not on stimulus properties might be not behaviorally relevant. Nonlinear effects might facilitate the encoding of faint signals during a three fish setting, the electrosensory cocktail party. The EOD frequencies of the encoutered three fish would be drawn from the EOD frequency distribution of these fish and a stable faint signal detection would require a response irrespective of the individual EOD frequencies. Weather integrating from a heterogeneous population with different \fbasesolid{} (50--450\,Hz, \citealp{Grewe2017, Hladnik2023}) would cover the behaviorally relevant range in the electrosensory cocktail party should be addressed in further studies. 
 
+%In this work, nonlinear effects were always found only for specific frequencies in relation to \fbase{}, corresponding to findings from previous literature \citealp{Voronenko2017}.
 %Only a heterogeneous population  could cover the whole stimulus space required during the electrosensory cocktail party.
 %If pyramidal cells would integrate only from P-units with the same mean baseline firing rate \fbasesolid{} not all fish encounters, relevant for the context of the electrosensory cocktail party, could be covered. 
 
@@ -642,7 +643,9 @@ A heterogeneous readout might be not only physiologically plausible but also req
 
 %These low-frequency modulations of the amplitude modulation are
 \subsection{Encoding of secondary envelopes}%($\n{}=49$)	Whether this population of envelope encoders was in addition bursty was not addressed in the corresponding study
-The RAM stimulus used in this work is an approximation of the three-fish scenario, where the two generated beats are often slowly modulated at the difference between the two beat frequencies \bdiff{}, known as secondary or social envelope \citealp{Stamper2012Envelope}. In previous works it was demonstrated that low-frequency secondary envelopes are extracted not in P-units but downstream of them in the ELL \citealp{Middleton2006} utilizing threshold nonlinear response curves of the involved neuron \citealp{Middleton2007}. Based on our work we would predict that only a small class of cells, with very low CVs, should encode the social envolope at the difference frequency. The sample in that previous work  \citealp{Middleton2007} was limited, and if it did not contain cells as low CVs as mentioned here, this might be in line with the findings that P-units were determined as no envelope encoders. 
+The RAM stimulus used in this work is an approximation of the three-fish scenario, where the two generated beats are often slowly modulated at the difference between the two beat frequencies \bdiff{}, known as secondary or social envelope \citealp{Stamper2012Envelope}. 
+
+In previous works it was demonstrated that low-frequency secondary envelopes are extracted not in P-units but downstream of them in the ELL \citealp{Middleton2006} utilizing threshold nonlinear response curves of the involved neuron \citealp{Middleton2007}. Based on our work we would predict that only a small class of cells, with very low CVs, should encode the social envelope at the difference frequency. If the sample in that previous work  \citealp{Middleton2007} did not contain low CVs cells, This could explain the conclusion that P-units were identified not as envelope encoders. 
 
 On the other hand in previous literature the encoding of social envelopes was attributed to a subpopulation of P-units with strong nonlinearities, low firing rates and high CVs \citealp{Savard2011}. These findings are in contrast to the findings in the previously mentioned work \citealp{Middleton2007} and on first glance also to our findings. The missing link, that has not been considered in this work, might be bursting of P-units, the repeated firing of spikes after one EOD period interleaved with quiescence (unpublished work). Bursting was not explicitly addressed in the work \citealp{Savard2011}, still the high CVs of the envelope encoding P-units indicate a higher rate of bursting. How bursts influence the second-order susceptibility of P-units will be addressed in following works (in preparation). 
 

utils_all_down.py

@@ -1678,7 +1678,6 @@ def do_withenoise_stimulus(deltat, eod_fr, stimulus_length, a_fe=0.2):
     stimulus_here[stimulus_here < 0] = 0
     return stimulus_here
 
-
 def model_sheme_split(grid_sheme_orig, time_transform=1000, ws=0.1, nfft=4096 * 6, stimulus_length=5, fft_type='mppsd',
                       a_fr=1, a_fe=0.2,
                       v_exp=1, exp_tau=0.1, counter=0, counterp=0, color='darkgrey', log=True,
@@ -1727,6 +1726,411 @@ def model_sheme_split(grid_sheme_orig, time_transform=1000, ws=0.1, nfft=4096 *
     # ult_settings(column=2, length=5)
     #     # fdefaig = plt.figure()
 
+    # for mult_nr in range(len(eod_fe)):
+    c_sigs = [1, 1, 1, 0.9]
+    c_sigs = [0, 0, 0, 0.9]
+    var_types = ['', '', '', 'additiv_cv_adapt_factor_scaled']  # ''#'additiv_cv_adapt_factor_scaled'
+    #            'additiv_visual_d_4_scaled', '']
+    # a_fes =
+    nrs = [1, 2, 3, 4]
+    a_fes = [0, 0.02, 0.1, 0]#alpha\alpha
+    titles = ['Baseline', r'Contrast$\,=2\,\%$', r'Contrast$\,=20\,\%$', noise_name()]
+
+    #########################################################################################
+    # first row for the stimulus, and then three cols for the sheme, and the power 1 and power 3
+    grid0 = gridspec.GridSpecFromSubplotSpec(1, len(a_fes) + 1, subplot_spec=grid_sheme_orig,
+                                             width_ratios=[1.5, 2, 2, 2, 2], wspace=0.45)
+
+    #####################################################
+    # Grid for the sheme
+    try:
+        grid_sheme = gridspec.GridSpecFromSubplotSpec(6, 1,
+                                                      subplot_spec=grid0[0], wspace=0.2, hspace=0.9,
+                                                      height_ratios=[1, 1, 1, 1, 0, 1])  # 0.95
+    except:
+        print('grid thing1')
+        embed()
+
+    axshemes = []
+    axsheme = plt.subplot(grid_sheme[0])
+    axshemes.append(axsheme)
+    plot_sheme_nonlinearity(axsheme, color_p3, color_p1, color_diagonal=color_diagonal)
+    axsheme = plt.subplot(grid_sheme[1])
+    axshemes.append(axsheme)
+    # axsheme.set_aspect('equal')
+    plot_sheme_lowpass(axsheme)
+
+    axsheme = plt.subplot(grid_sheme[2])
+    axshemes.append(axsheme)
+    plot_sheme_noise(axsheme)
+
+    axsheme = plt.subplot(grid_sheme[3])
+    axshemes.append(axsheme)
+    # axsheme.set_aspect('equal')
+    plot_sheme_IF(axsheme, exp_tau, v_exp)
+
+    ###################################################################################
+
+    lw = 0.5
+
+    xlim = 0.065
+    axps = []
+    axps_lowpass = []
+    axps_stimulus = []
+    pps = []
+    pps_lowpass = []
+    pps_stimulus = []
+    ax_am_sp_s = []
+    ax_ams = []
+    ax_noise = []
+    colors_chosen = []
+    counter_g = 0
+    mult_nr = 0
+
+    # A grid for a single POWER column
+
+    axt_stims = []
+
+    for c, c_sig in enumerate(c_sigs):
+
+        a_fe = a_fes[c]
+
+        ######################################
+        # colors = ['grey', 'grey', 'grey', 'grey']
+        # embed()
+        grid_power_col = gridspec.GridSpecFromSubplotSpec(6, 1,
+                                                          subplot_spec=grid0[nrs[counter_here]], wspace=0.45,
+                                                          hspace=0.5, height_ratios=[1, 1, 1, 1, 0, 1])
+
+        noise_final_c, spike_times, stimulus, stimulus_here, time, v_dent_output, v_mem_output,frame  = get_flowchart_params(
+            a_fes, a_fr, c, c_sig, cell, deltat, eod_fr, model_params, stimulus_length, v_offset, var_types, eod_fe,
+            color_p1, color_p3, mult_nr=mult_nr, load_name=load_name, exp_tau=exp_tau, v_exp=v_exp)
+        print(len(stimulus_here))
+        ##############################################
+        # titles = [titles[1]]
+        # for g, stimulus_here in enumerate([stimuli[1]]):
+        add = 0
+        color = colors[counter_here]
+
+        # FIRST Row: Rectified stimulus
+
+        plot_point = [[], [], [], 'yes']
+        # counter, ax_rec[counter_here], ff, pp, axp = plot_rec_stimulus(eod_fr,grid_lowpass, time_transform, stimulus_here, color, color,
+        #                            time, counter, eod_fr, titles, g, deltat,
+        #                            fft_type, nfft, log, counterp,shift = shift, delta_f  = delta_f,  plot_point =   plot_point[counter_here], lw = lw, xlim = xlim)
+        power_extra = False
+        wr2 = [1, 1.2]
+        if power_extra:
+            wr = [1, 1.2]
+
+            col = 2
+        else:
+            col = 1
+            wr = [1]
+
+        ####################################################################
+        # stimulus
+        grid_lowpass = gridspec.GridSpecFromSubplotSpec(1, col,
+                                                        subplot_spec=grid_power_col[0], wspace=ws, hspace=1.3,
+                                                        width_ratios=wr)
+        ax_rec[counter_here], ff, pp, ff_am, pp_am = plot_lowpass2(grid_lowpass, time, stimulus_here, deltat, eod_fr,
+                                                                   shift, nfft, time_transform, color, fft_type, lw=lw,
+                                                                   xlim=xlim)
+
+        # ax_rec[counter.]
+        ax_rec[counter_here].show_spines('b')
+        remove_xticks(ax_rec[counter_here])
+        axt_stims.append(ax_rec[counter_here])
+
+        pps_stimulus.append(pp_am)
+
+        if power_extra:
+            axp_p2 = plt.subplot(grid_lowpass[1])
+            axp_p2.set_xticks_blank()
+            axps_stimulus.append(axp)
+        colors_chosen.append(color)
+
+        # if counter_here == 0:
+        #    ax_rec[counter_here].text(-7, 0, '0', color='black', ha='center', va='center')
+
+        # if colorful_title:
+        #    rainbow_title(fig, ax_rec[counter_here], titles[g], add_pos[g], color_add_pos[g])
+        # else:#add_pos[g]
+        #embed()
+        ax_rec[counter_here].text(0, 0.935, titles[c] + '\n'+r'$\rm{CV}=%s$'  %(
+            np.round(np.std(np.diff(spike_times)) / np.mean(np.diff(spike_times)), 2)) + '\n$f_{Base}=%s$' %(
+            int(np.round(1 / np.mean(np.diff(spike_times))))) + '\,Hz', transform=ax_rec[counter_here].transAxes,
+                                  va='bottom')  # verticalalignment='right',
+        # And plot correspoding sheme
+        # if g == 0:
+
+        # REMAINING Rows: dendridic filter / LIF /EIF stimulus
+
+        # for ee, exponential in enumerate(exponentials):
+
+        # model
+        # v_offset,  model_params, load_name= implement_three_core(cell,amp_frame, titles, g, cell_nr = cell_nr)
+        # for hard coding the offset here i check the change of the baseline
+
+        # if (ee == 0):
+        # SECOND Row: Dendridic Low pass filter
+        plot_point = [[], [], [], 'yes']
+        # ax_low[counter_here]
+        grid_lowpass = gridspec.GridSpecFromSubplotSpec(1, col,
+                                                        subplot_spec=grid_power_col[1], wspace=ws, hspace=1.3,
+                                                        width_ratios=wr)
+        axt2, ff, pp, ff_am, pp_am = plot_lowpass2(grid_lowpass, time, v_dent_output, deltat, eod_fr, shift, nfft,
+                                                   time_transform, color, fft_type, lw=lw, xlim=xlim)
+        axt2.show_spines('b')
+        remove_xticks(axt2)
+        axt_stims.append(axt2)
+
+        pps_stimulus.append(pp_am)
+        if power_extra:
+            axp_p2 = plt.subplot(grid_lowpass[1])
+            axp_p2.set_xticks_blank()
+            axps_stimulus.append(axp_p2)
+        colors_chosen.append(color)
+
+        # if counter_here == 0:
+        #    ax_low[counter_here].text(-7, 0, '0', color='black', ha='center', va='center')
+
+        ####################################################################
+        # spikes
+
+        # embed()
+        grid_lowpass = gridspec.GridSpecFromSubplotSpec(1, 1,
+                                                        subplot_spec=grid_power_col[-3], wspace=ws, hspace=1.3,
+                                                        )  # width_ratios=wr2
+
+        # add = adds[g][2+ee]
+        plot_point = ['yes', 'yes', [], 'yes']
+        # embed()
+        axt_spikes, axp_IF, ff, pp, axp_s, pp_s = plot_spikes(grid_lowpass, time_transform, v_mem_output, time, color,
+                                                              spike_times, shift, deltat, fft_type, nfft, eod_fr,
+                                                              xlim=xlim,
+                                                              counter_here=counter_here, psd=False)  # , add = add
+
+        grid_lowpass_p = gridspec.GridSpecFromSubplotSpec(1, 1,
+                                                          subplot_spec=grid_power_col[-1], wspace=ws, hspace=1.3,
+                                                          )  # width_ratios=wr2
+        axp_s = plt.subplot(grid_lowpass_p[0])
+
+        in_omega = False
+        if in_omega:
+            axp_s.set_xlabel('$\omega/\omega_1$')
+        else:
+
+            arrow = True
+            name = r'$f/f_{EOD}$'#r'$\frac{f}{f_{EOD}}$'
+            if arrow:
+                set_xlabel_arrow_core(axp_s, name )
+                #axp_s.text(1.05, -0.25, name, ha='center', va='center',
+                #           transform=axp_s.transAxes)
+                axp_s.arrow_spines('b')
+            else:
+                axp_s.set_xlabel(name)
+
+
+        # pps.append(pp)
+        # axps.append(axp_IF)
+        axps.append(axp_s)
+        pps.append(pp_s)
+        # colors_chosen.append(color)
+        colors_chosen.append('black')
+        # if ee == 0:
+
+        axt_spikes.show_spines('b')
+        axt_IF1.append(axt_spikes)
+        # axt_stims,axt_IF1
+        # else:
+        #    axt_IF2.append(axt_IF)
+        # if g == 0:
+
+        ################################
+        # plot noise split
+        noise = np.random.randn(len(stimulus))
+        # noise *= noise_strength / np.sqrt(deltat)
+        noise_strength = model_params["noise_strength"]  # .iloc[0]
+        # if 'd_right' in load_name:
+        noise_length = len(stimulus)
+
+        # noise_final = np.random.randn(noise_length)  # len(stimulus)
+        # noise_strength_new = np.sqrt(noise_strength * 2)
+        # noise_final *= noise_strength_new / np.sqrt(deltat)  # 0.05370289258320868 0.0015532069917408744
+
+        no_noise = False
+        if no_noise:
+            c_noise = c_sig
+            noise_final_c = np.random.randn(noise_length)  # len(stimulus)
+            variance = (noise_strength * c_noise) * 2 / deltat
+            noise_strength_new = np.sqrt(variance)
+            noise_final_c *= noise_strength_new  # 0.0015532069917408744
+        grid_lowpass = gridspec.GridSpecFromSubplotSpec(1, col, width_ratios=wr,
+                                                        subplot_spec=grid_power_col[2], wspace=ws,
+                                                        hspace=1.45)
+
+        ax_n, ff, pp, ff_am, pp_am = plot_lowpass2(grid_lowpass, time, noise_final_c, deltat, eod_fr, shift, nfft,
+                                                   time_transform, color, fft_type, extract=False, lw=lw, xlim=xlim)
+        ax_n.show_spines('b')
+        ax_noise.append(ax_n)
+        remove_xticks(ax_n)
+        pps_lowpass.append(pp)
+        if power_extra:
+            axp_p2 = plt.subplot(grid_lowpass[1])
+            axp_p2.set_xticks_blank()
+            axps_lowpass.append(axp_p2)
+        # pps_stimulus.append(pp)
+        # axps_stimulus.append(axp_p2)
+        # axps.append(axp_p2)
+        # pps.append(pp)
+
+        counter_g += 1
+        # embed()
+        # plt.show()
+        counter_here += 1
+
+        # embed()
+
+
+    # embed()
+    devide = np.max(np.max(pps))
+    # plot_points = np.array([[], [], [], 'yes',
+    #                       [], [], [], 'yes'
+    #                     ,'yes','yes',[],'yes'
+    #                ,'yes','yes',[],'yes'])
+    # von oben nach unten von links nach rechts
+
+    # plot psd with shared log lim
+
+    ####################################
+    # cut first parts
+    # because otherwise there is a dip at the beginning and thats a problem for the range thing
+    ff, pps_stimulus, pps_lowpass, pps = cut_first_parts(ff, pps_stimulus, pps_lowpass, pps, ll=0)
+
+    # here I calculate the log and do the same range for all power spectra
+    # this is kind of complicated but some cells spike even withouth thresholding and we want to keep their noise floor down
+    # not to see the peaks in the noise
+    # pp3_stimulus = create_same_max(np.concatenate([pps_stimulus, pps_lowpass]), same=True)
+
+    if power_extra:
+        pp3_stimulus = create_same_max(pps_stimulus, same=True)
+        pp3_noise = create_same_max(pps_lowpass, same=True)
+        pps3 = create_same_max(pps, same=True)
+
+        # pp3_stimulus = np.concatenate(pp3_stimulus)
+        # axps_stimulus = np.concatenate([axps_stimulus,axps_lowpass])
+        # pp3_stimulus = create_same_range_ps(axps, pps_stimulus)
+
+        pp3 = create_same_range(np.concatenate([pp3_stimulus, pp3_noise, pps3]))
+        axps_stimulus = np.concatenate([axps_stimulus, axps_lowpass, axps])
+    else:
+        axps_stimulus = axps
+        pp3 = pps
+    # embed()
+
+    # pp3[4] = np.array([float('nan')]*len(pp3[4]))
+    # embed()
+    # ok i do this because most cells dont spike, but they are not nice to show because of the very high offset between persynaptci potenatil and spike
+    # there are only few cells where the distance is not so high and this cells spike occationally but very randomly still we dont wanna se their power specturm
+    # therefore we dont show it
+    colors = [color_diagonal, color_p1, color_p1, color_p3,
+              color_diagonal, color_p1, color_p1, color_p3,
+              color_diagonal, color_p1, color_p1, color_p3, ]
+
+    plot_points = ['yes', 'yes', 'yes', 'yes',
+                   'yes', 'yes', 'yes', 'yes',
+                   'yes', 'yes', 'yes', 'yes',
+                   'yes', 'yes', 'yes', 'yes', ]
+    plot_points = [[], 'yes', [], 'yes',
+                   [], 'yes', [], 'yes',
+                   [], 'yes', [], 'yes',
+                   [], 'yes', [], 'yes', ]
+
+    axt_stims[0].get_shared_y_axes().join(*axt_stims)
+    axt_IF1[0].get_shared_y_axes().join(*axt_IF1)
+    ax_noise[0].get_shared_y_axes().join(*ax_noise)
+    set_same_ylim(ax_noise)
+    # embed()
+    ax = np.transpose([axt_stims[0::2], axt_stims[1::2],ax_noise,  axt_IF1, axps_stimulus])
+
+    fig = plt.gcf()
+    #fig.tag(ax, xoffs=-3.5, yoffs = 1.5)
+    nr = 1.7#2#.24
+    tag2(fig, axshemes, xoffs=-3.5, yoffs=5.5)
+    tag2(fig, np.transpose(ax), xoffs=-3.5)#yoffs = [5.5,nr,nr,nr,nr-0.2]
+    # get_ylim_same
+
+    #########################################################
+    # plot psds
+    for a, axp in enumerate(axps_stimulus):
+        lw_p = 0.8
+        # gemeinsamen Limit setzten
+        pp_here = 2*10 * np.log10(pp3[a] / np.max(pp3))# der faktor 2 ist falsch
+        plot_power_common_lim(axp, pp_here, 0, ff / eod_fr, colors[a], lw_p, plot_points[a], delta_f / eod_fr,
+                              log=False,
+                              devide=devide)
+
+        axp.show_spines('b')
+        axp.set_ylim(-22, 1)  # -8.5
+        if a == 3:  # % 4 == 3:
+            axp.yscalebar(1.1, 0.6, 20, 'dB', va='center', ha='right')
+        #axp.yscalebar(1.1, 0.8, 20, 'dB', va='center', ha='right')
+        #embed()
+
+    axps_stimulus[0].get_shared_y_axes().join(*axps_stimulus)
+    # if power_extra:
+    #    axps_stimulus[0].get_shared_y_axes().join(*axps_stimulus)
+
+
+def model_sheme_split2(grid_sheme_orig, time_transform=1000, ws=0.1, nfft=4096 * 6, stimulus_length=5, fft_type='mppsd',
+                      a_fr=1, a_fe=0.2,
+                      v_exp=1, exp_tau=0.1, counter=0, counterp=0, color='darkgrey', log=True,
+                      shift=0.25):
+    load_name = 'models_big_fit_d_right.csv'
+    cell_nr = 8  # 5#5#6#3
+    # model_params = load_model(load_name=load_name, cell_nr = cell_nr)
+    models = resave_small_files("models_big_fit_d_right.csv", load_folder='calc_model_core')
+    flowchart_cell = '2012-07-03-ak-invivo-1'
+    model_params = models[models['cell'] == flowchart_cell].iloc[0]
+    print('cell='+str(flowchart_cell))
+    # amp_frame = pd.read_csv('peak_amplitudes_power.csv')
+
+    cell = model_params.pop('cell')  # .iloc[0]# Werte für das Paper nachschauen
+    eod_fr = model_params['EODf']  # .iloc[0]
+    deltat = model_params.pop("deltat")  # .iloc[0]
+    v_offset = model_params.pop("v_offset")  # .iloc[0]
+
+    sampling_rate = 1/deltat
+    nfft = 2**20#int(sampling_rate / 4)
+    #embed()
+    cell_nr = 8  # 5#5#6#3
+    # embed()
+
+    eod_fe = [eod_fr + 50]  # eod_fr*1+50,, eod_fr * 2 + 50
+    mult_nr = 0
+
+    # REMAINING rows
+    color_p3 = 'grey'  # 'red'#palette['red']
+    color_p1 = 'grey'  # 'blue'#palette['blue']
+    color_diagonal = 'grey'  # 'cyan'#palette['cyan']
+    colors = [color_diagonal, color_p1, color_p1, color_p3]
+
+    ax_rec = [[]] * 4
+    ax_n = [[]] * 4
+    axt_IF1 = []
+    axt_IF2 = []
+    adds = [[0, 0, 0, 0], [0, 0, 2, 10]]
+    nrs_grid = [0, 1, 3, 4]
+    # delta_f = (eod_fe[mult_nr] - eod_fr) - eod_fr
+    delta_f = [50]  # create_beat_corr(np.array([eod_fe[mult_nr] - eod_fr]), np.array([eod_fr]))[0]
+
+    # time, stimulus_here, eod_fish_r, eod_fish_e, stimulus = make_paramters(
+    #    stimulus_length, deltat, eod_fr, a_fr, a_fe, eod_fe, mult_nr)
+    counter_here = 0
+    # ult_settings(column=2, length=5)
+    #     # fdefaig = plt.figure()
+
     # for mult_nr in range(len(eod_fe)):
     c_sigs = [1, 1,  0.9]#1,
     c_sigs = [0, 0,  0.9]#'',0,