Reporting last changes from @pompolas PR #13

#13
mpompolas · Apr 29, 2022 · d55a3b8 · d55a3b8
1 parent ae9cd96
commit d55a3b8
Show file tree

Hide file tree

Showing 7 changed files with 21 additions and 15 deletions.
diff --git a/toolbox/process/functions/process_convert_raw_to_lfp.m b/toolbox/process/functions/process_convert_raw_to_lfp.m
@@ -172,7 +172,7 @@
     % Template structure for the creation of the output raw file
     sFileTemplate = sFileIn;
     sFileTemplate.prop.sfreq = LFP_fs;
-    sFileTemplate.prop.times = [0, (nTimeOut-1) ./ LFP_fs];
+    sFileTemplate.prop.times = [newTimeVector(1), newTimeVector(end)];
     sFileTemplate.header.sfreq = LFP_fs;
     sFileTemplate.header.nsamples = nTimeOut;
     % Convert events to new sampling rate
@@ -231,7 +231,8 @@
         % Get channel name from electrode file name
         [tmp, ChannelName] = fileparts(ElecFile);
         ChannelName = strrep(ChannelName, 'raw_elec_', '');
-        data = BayesianSpikeRemoval(ChannelName, data, sr, sFileIn, ChannelMat, cleanChannelNames);
+        data = BayesianSpikeRemoval(ChannelName, data, sr, sFileIn, ChannelMat, cleanChannelNames, BandPass);
+        data = data';
     end
     % Band-pass filter
     data = bst_bandpass_hfilter(data, sr, BandPass(1), BandPass(2), 0, 0);
@@ -242,7 +243,7 @@
 
 %% ===== BAYESIAN SPIKE REMOVAL =====
 % Reference: https://www.ncbi.nlm.nih.gov/pubmed/21068271
-function data_derived = BayesianSpikeRemoval(ChannelName, data, Fs, sFile, ChannelMat, cleanChannelNames)
+function data_derived = BayesianSpikeRemoval(ChannelName, data, Fs, sFile, ChannelMat, cleanChannelNames, BandPass)
     % Assume that a spike lasts 3ms
     nSegment = round(Fs * 0.003);
     Bs = eye(nSegment); % 60x60
@@ -275,7 +276,7 @@
         % from spktimes to obtain the start times of the spikes
 
         if mod(length(data),2)~=0
-            data_temp = [data;0];
+            data_temp = [data 0]';
             g = fitLFPpowerSpectrum(data_temp,BandPass(1),BandPass(2),sFile.prop.sfreq);
             S = zeros(length(data_temp),1);
             iSpk = round(spkSamples - nSegment/2);

diff --git a/toolbox/process/functions/process_noise_correlation.m b/toolbox/process/functions/process_noise_correlation.m
@@ -78,7 +78,7 @@
         end
     end
     % If no neuron was found
-    if isempty(neuronLabels)
+    if isempty(uniqueNeurons)
         bst_report('Error', sProcess, sCurrentInputs(1), 'No neurons/spiking events detected.');
         return;
     end

diff --git a/...ss/functions/process_psth_per_electrode.m → ...cess/functions/process_psth_per_channel.m b/...ss/functions/process_psth_per_electrode.m → ...cess/functions/process_psth_per_channel.m
@@ -1,8 +1,8 @@
-function varargout = process_psth_per_electrode( varargin )
-% PROCESS_PSTH_PER_ELECTRODE: Computes the PSTH per electrode.
+function varargout = process_psth_per_channel( varargin )
+% PROCESS_PSTH_PER_CHANNEL: Computes the PSTH per channel.
 
-% It displays the binned firing rate on each electrode (of only the first 
-% neuron on each electrode if multiple have been detected). This can be nicely
+% It displays the binned firing rate on each channel (of only the first 
+% neuron on each channel if multiple have been detected). This can be nicely
 % visualized on the cortical surface if the positions of the electrodes
 % have been set, and show real time firing rate.
 
@@ -34,7 +34,7 @@
 %% ===== GET DESCRIPTION =====
 function sProcess = GetDescription()
     % Description the process
-    sProcess.Comment     = 'PSTH per electrode';
+    sProcess.Comment     = 'PSTH per channel';
     sProcess.FileTag     = 'raster';
     sProcess.Category    = 'File';
     sProcess.SubGroup    = 'Electrophysiology';

diff --git a/toolbox/process/functions/process_psth_per_neuron.m b/toolbox/process/functions/process_psth_per_neuron.m
@@ -92,7 +92,7 @@
             end
         end
         % If no neuron was found
-        if isempty(neuronLabels)
+        if isempty(labelsNeurons)
             bst_report('Error', sProcess, sCurrentInputs(1), 'No neurons/spiking events detected.');
             return;
         end

diff --git a/toolbox/process/functions/process_rasterplot_per_neuron.m b/toolbox/process/functions/process_rasterplot_per_neuron.m
@@ -82,7 +82,7 @@
             end
         end
         % If no neuron was found
-        if isempty(neuronLabels)
+        if isempty(labelsNeurons)
             bst_report('Error', sProcess, sCurrentInputs(1), 'No neurons/spiking events detected.');
             return;
         end

diff --git a/toolbox/process/functions/process_spike_triggered_average.m b/toolbox/process/functions/process_spike_triggered_average.m
@@ -165,7 +165,7 @@
             end 
         end
         % Divide by total number of averages
-        STA_single_neuron = (STA_single_neuron./divideBy)';
+        STA_single_neuron = (STA_single_neuron./divideBy);
 %         std_single_neuron = sqrt(std_single_neuron./(divideBy - size(all_labels,2)));
 
         % Get meaningful label from neuron name
@@ -179,7 +179,7 @@
         % ===== SAVE FILE =====
         % Prepare output file structure
         FileMat = db_template('datamat');
-        FileMat.F           = STA_single_neuron';
+        FileMat.F           = STA_single_neuron;
         FileMat.Time        = time_segmentAroundSpikes; 
 %         FileMat.Std         = 2 .* std_single_neuron; % MULTIPLY BY 2 TO GET 95% CONFIDENCE (ASSUMING NORMAL DISTRIBUTION)
         FileMat.Comment     = ['Spike Triggered Average: ' str_remove_parenth(DataMats{1}.Comment) ' (' better_label ')'];

diff --git a/toolbox/process/functions/process_spikesorting_kilosort.m b/toolbox/process/functions/process_spikesorting_kilosort.m
@@ -496,7 +496,12 @@ function ImportKilosortEvents(sFile, ChannelMat, parentPath, rez)
             index = index + 1;
         end
     end
-    events = [events events_spikes];
+
+    if ~isempty(existingEvents)
+        events = [events events_spikes];
+    else
+        events = events_spikes;
+    end
 
     save(fullfile(parentPath,'events_UNSUPERVISED.mat'),'events')