Added includeFlag for FilterRecordings to use that column from recording database, findMRCs fixes integration units and ensures nReps/distance will work even if other output params are added in between, multiple columns and usability for tau export to Igor for AntVPost_voltage attenuation script, IdAnalysis fixes roundVel using correct column, FreqAnalysis fixes column

Author: sammykatta

Analysis/SK/AnalyzePatchData.m

@@ -60,12 +60,13 @@
 cellDist = [40 200];
 resistCutoff = '<250';
 extFilterFreq = 2.5;
+includeFlag = 1;
 
 anteriorDistCells = FilterRecordings(ephysData, ephysMetaData,...
     'strain', strainList, 'internal', internalList, ...
      'stimLocation', stimPosition, 'wormPrep', wormPrep, ...
      'cellStimDistUm',cellDist, 'RsM', resistCutoff, ...
-     'stimFilterFrequencykHz', extFilterFreq);
+     'stimFilterFrequencykHz', extFilterFreq, 'included', 1);
 
 clear cellDist strainList internalList cellTypeList stimPosition resistCutoff ans wormPrep;
 

Analysis/SK/AnnotatedScripts/AntVsPost_ToIgor_180911.m

@@ -0,0 +1,308 @@
+% Tau vs distance curves for anterior vs. posterior
+
+% Running all the way through allows user to pull out tau data from steps, 
+% export to Igor, fit a Boltzmann there, re-import and normalize and export
+% back to Igor for final plotting and fitting. 
+
+% Unlike AntVsPost_VoltageAttenuation, this does not correct the current or
+% charge for voltage errors based on distance, and was used for non-current
+% data.
+
+%%Import data and metadata
+ephysData = ImportPatchData('incl',1);
+projects = {'FAT';'SYM'};
+ephysData = FilterProjectData(ephysData, projects);
+clear projects;
+
+ephysMetaData = ImportMetaData();  %Recording Database.xlsx
+attenuationData = ImportMetaData(); %AttenuationCalcs.xlsx
+
+%% Analyze capacity transient for C, Rs, and tau
+
+ephysData = CtAnalysis(ephysData);
+
+clear lastfit;
+
+%% Select recordings that match parameters based on metadata
+
+strainList = {'TU2769'};
+internalList = {'IC2'};
+stimPosition = {'posterior'};
+
+wormPrep = {'dissected'};
+cellDist = [40 90]; % stimulus/cell distance in um
+resistCutoff = '<250'; % Rs < 250 MOhm
+extFilterFreq = 2.5; % frequency of low-pass filter for stimulus command
+includeFlag = 1; 
+
+posteriorDistCells = FilterRecordings(ephysData, ephysMetaData,...
+    'strain', strainList, 'internal', internalList, ...
+    'stimLocation', stimPosition, 'wormPrep', wormPrep, ...
+    'cellStimDistUm',cellDist, 'RsM', resistCutoff, ...
+    'stimFilterFrequencykHz', extFilterFreq, 'included', includeFlag);
+
+stimPosition = {'anterior'};
+
+anteriorDistCells = FilterRecordings(ephysData, ephysMetaData,...
+    'strain', strainList, 'internal', internalList, ...
+    'stimLocation', stimPosition, 'wormPrep', wormPrep, ...
+    'cellStimDistUm',cellDist, 'RsM', resistCutoff, ...
+    'stimFilterFrequencykHz', extFilterFreq, 'included',includeFlag);
+
+
+clear cellDist strainList internalList cellTypeList stimPosition resistCutoff ans wormPrep excludeCells;
+
+% This results in a file with three columns: cell ID, series, and sweeps 
+% that have been approved for analysis use.
+
+%% Visual/manual exclusion of bad sweeps 
+% This is mainly based on excluding sweeps with leak > 10pA, but also
+% sweeps where the recording was lost partway through or some unexpected
+% source of noise was clearly at play.
+
+protList ={'WC_Probe';'WC_ProbeSmall';'WC_ProbeLarge';'NoPrePulse'};
+matchType = 'full';
+
+ExcludeSweeps(ephysData, protList, anteriorDistCells, 'matchType', matchType);
+ExcludeSweeps(ephysData, protList, posteriorDistCells, 'matchType', matchType);
+
+%% Find MRCs 
+% antAllSteps.xlsx and postAllSteps.xlsx from 180810
+protList ={'WC_Probe';'NoPre'};
+matchType = 'first';
+
+sortSweeps = {'magnitude','magnitude','magnitude','magnitude'};
+
+anteriorMRCs = IdAnalysis(ephysData,protList,anteriorDistCells,'num','matchType',matchType, ...
+    'tauType','thalfmax', 'sortSweepsBy', sortSweeps, 'integrateCurrent',1 , ...
+    'recParameters', ephysMetaData,'sepByStimDistance',1);
+
+posteriorMRCs = IdAnalysis(ephysData,protList,posteriorDistCells,'num','matchType',matchType, ...
+    'tauType','thalfmax', 'sortSweepsBy', sortSweeps, 'integrateCurrent',1 , ...
+    'recParameters', ephysMetaData,'sepByStimDistance',1);
+
+clear protList sortSweeps matchType
+
+%% Pull out and combine relevant data for plot
+
+% Voltage attenuation data comes from the length constant fitting done in
+% Igor, which gives a voltage attenuation factor at the location of the
+% stimulus site for each recording, based on the calculated length
+% constant.
+
+% Grab the current at the selected step size (e.g., 10um) for each
+% recording, grab the cell-stimulus distance for that recording, and match
+% the recording name against the voltage attenuation table to find the
+% attenuation factor for that recording.
+whichMRCs = anteriorMRCs;
+thisAtt = attenuationData(:,[2 8 10]);
+distCol = 12;
+peakCol = 8;     % 8 for tau act, 9 for tau decay
+distVPeak = [];
+stepSize = 10;
+
+for iCell = 1:size(whichMRCs,1)
+    thisCell = whichMRCs{iCell,3};
+    whichStep = round(thisCell(:,1)) == stepSize;
+    if any(whichStep)
+        thisName = whichMRCs{iCell,1};
+        hasAtt = strcmp(thisName,thisAtt(:,1));
+        
+        if any(hasAtt) && thisAtt{hasAtt,2}
+            distVPeak(iCell,:) = [thisCell(whichStep,[distCol peakCol]) thisAtt{hasAtt,3}];
+        else
+            distVPeak(iCell,:) = [thisCell(whichStep,[distCol peakCol]) nan];
+        end
+        
+    end
+    
+end
+
+distVPeak_Ant = distVPeak;
+
+whichMRCs = posteriorMRCs;
+distVPeak = [];
+
+
+for iCell = 1:size(whichMRCs,1)
+    thisCell = whichMRCs{iCell,3};
+    whichStep = round(thisCell(:,1)) == stepSize;
+    if any(whichStep)
+        thisName = whichMRCs{iCell,1};
+        hasAtt = strcmp(thisName,thisAtt(:,1));
+        
+        if any(hasAtt)
+            distVPeak(iCell,:) = [thisCell(whichStep,[distCol peakCol]) thisAtt{hasAtt,2}];
+        else
+            distVPeak(iCell,:) = [thisCell(whichStep,[distCol peakCol]) nan];
+        end
+        
+    end
+    
+end
+
+distVPeak_Post = distVPeak;
+
+clear a iCell thisCell whichMRCs whichStep thisName hasAtt distVPeak
+
+%% Export for Igor fitting of Boltzmann to each recording
+
+% This is the same as in AntVPost_VoltageAttenuation_180803 except it
+% doesn't apply any attenuation correction. Useful for pulling out the taus
+
+eachSize = [0.5 1 1.5 3 4 5 6 7 8 9 10 11 12]';
+distLimits = [40 90];
+% limit to same average distance for anterior and posterior
+
+whichMRCs = anteriorMRCs;
+thisAtt = attenuationData(:,[2 8 10]);
+thisName = cell(0);
+ant_Out = cell(length(eachSize)+1,size(whichMRCs,1));
+
+for iCell = 1:size(whichMRCs,1)
+    thisCell = whichMRCs{iCell,3};
+    thisDist = mean(thisCell(:,distCol)); % check if cell distance is in range
+    if thisDist <= distLimits(2) && thisDist >= distLimits(1)
+        thisName{iCell,1} = whichMRCs{iCell,1}; %name
+        thisName{iCell,2} = thisDist;
+        ant_Out{1,iCell} = thisName{iCell,1};
+        Iact = [];
+        
+        for iSize = 1:length(eachSize)
+            stepSize = eachSize(iSize);
+            whichStep = round(thisCell(:,1)*2)/2 == stepSize; %round to nearest 0.5
+            if any(whichStep)
+                
+                Iact(iSize,1) = thisCell(whichStep,peakCol);
+            else
+                Iact(iSize,1) = nan;
+            end
+        end
+        ant_Out(2:length(eachSize)+1,iCell) = num2cell(Iact);
+    end
+end
+ant_Name = [{'Anterior', 'Ant_Dist'}; thisName];
+ant_Out = ant_Out(:,~cellfun(@isempty, ant_Out(1,:))); % clear out empty waves (where dist didn't match)
+ant_Name = ant_Name(~cellfun(@isempty, ant_Name(:,1)),:);
+
+ant_Out = [[{'stepSize'};num2cell(eachSize)] ant_Out]; % append stepSize wave
+
+thisName = cell(0);
+whichMRCs = posteriorMRCs;
+post_Out = cell(length(eachSize)+1,size(whichMRCs,1));
+
+for iCell = 1:size(whichMRCs,1)
+    thisCell = whichMRCs{iCell,3};
+    thisDist = mean(thisCell(:,distCol)); % check if cell distance is in range
+    if thisDist <= distLimits(2) && thisDist >= distLimits(1)
+        thisName{iCell,1} = whichMRCs{iCell,1}; %name
+        thisName{iCell,2} = thisDist; %distance
+        post_Out{1,iCell} = thisName{iCell,1};
+        Iact = [];
+        
+        for iSize = 1:length(eachSize)
+            stepSize = eachSize(iSize);
+            whichStep = round(thisCell(:,1)*2)/2 == stepSize; %round to nearest 0.5
+            if any(whichStep)
+                Iact(iSize,1) = thisCell(whichStep,peakCol);
+            else
+                Iact(iSize,1) = nan;
+            end
+        end
+        post_Out(2:length(eachSize)+1,iCell) = num2cell(Iact);
+    end
+end
+
+post_Name = [{'Posterior', 'Post_Dist'}; thisName];
+post_Out = post_Out(:,~cellfun(@isempty, post_Out(1,:)));
+post_Name = post_Name(~cellfun(@isempty, post_Name(:,1)),:);
+
+% Set the filename
+fname = 'PatchData/taus_antVpost_distLim(180911).xls';
+
+xlswrite(fname,ant_Out,'antAct');
+xlswrite(fname,post_Out,'postAct');
+xlswrite(fname,ant_Name,'antActStats');
+xlswrite(fname,post_Name,'postActStats');
+
+
+clear iCell thisCell Iact whichMRCs whichStep hasAtt thisAtt Vc Ena thisName
+
+
+%% Renormalize
+
+% The point of the remaining code is to grab the data from Igor or from an Excel
+% sheet after fitting individual I-d curves to get max/delta/xhalf from the
+% Boltzmann fit, then normalize each I-d curve to its predicted max, and
+% export back into an Igor-readable format for plotting.
+
+% From AttCorrectedID_Curve.m
+
+% From Igor, export the entire Igor file as a csv
+%% Import fits from Igor table
+
+fitStats = ImportMetaData(); %AttCorrectedI-D_fitstats.xlsx
+fitHeaders = fitStats(1,:);
+fitStats = fitStats(2:end,:);
+
+%% Get names and match up stats
+antStats = fitStats(:,cell2mat(cellfun(@(x) ~isempty(regexp(x,'Ant')), fitHeaders,'un',0)));
+postStats = fitStats(:,cell2mat(cellfun(@(x) ~isempty(regexp(x,'Post')), fitHeaders,'un',0)));
+antMaxIdx = find(~cellfun(@isempty,(regexp(fitHeaders(:,cell2mat(cellfun(@(x) ~isempty(regexp(x,'Ant')), fitHeaders,'un',0))),'[mM]ax'))));
+postMaxIdx = find(~cellfun(@isempty,(regexp(fitHeaders(:,cell2mat(cellfun(@(x) ~isempty(regexp(x,'Post')), fitHeaders,'un',0))),'[mM]ax'))));
+antStats = antStats(cellfun(@(x) ~isnumeric(x), antStats(:,1)),:);
+postStats = postStats(cellfun(@(x) ~isnumeric(x), postStats(:,1)),:);
+
+% this gives five columns: name, delta, distance, max, xhalf
+
+% straight up taking the fits and recoloring them here won't work because 
+% Igor fits are all 200 points long but the X wave
+% is calculated, and is different length for each (i.e., some curves have
+% X values from 0.5 to 11, some from 3 to 10, etc., and those ranges
+% determine what the x values are for each fit separately).
+
+% instead, I'm just going to do the normalization here and send the waves
+% back to Igor, where I can average them like the other I-dCellFits
+
+%% 
+eachSize = fitStats(:,cell2mat(cellfun(@(x) ~isempty(regexp(x,'stepSize')), fitHeaders,'un',0)));
+eachSize = eachSize(cellfun(@(x) isnumeric(x) && ~isnan(x),eachSize));
+
+whichSide = antStats;
+thisCell = whichSide(:,1);
+thisDataNorm = cell(0);
+
+for i = 1:size(whichSide,1)
+   thisMax = whichSide{i,antMaxIdx};
+   whichData = cell2mat(cellfun(@(x) ~isempty(regexp(x,sprintf('^(?!fit).*%s',thisCell{i}))),fitHeaders,'un',0));
+   thisData = fitStats(1:length(eachSize),whichData);
+   thisData(cellfun(@isempty,thisData))={nan};
+   thisDataNorm(:,i) = cellfun(@(x) x./thisMax, thisData,'un',0);
+end
+
+antNorm = thisDataNorm;
+antHeaders = cellfun(@(x) sprintf('%s_Norm',x),thisCell,'un',0)';
+
+
+% POSTERIOR
+whichSide = postStats;
+thisCell = whichSide(:,1);
+thisDataNorm = cell(0);
+
+for i = 1:size(whichSide,1)
+   thisMax = whichSide{i,postMaxIdx};
+   whichData = cell2mat(cellfun(@(x) ~isempty(regexp(x,sprintf('^(?!fit).*%s',thisCell{i}))),fitHeaders,'un',0));
+   thisData = fitStats(1:length(eachSize),whichData);
+   thisData(cellfun(@isempty,thisData))={nan};
+   thisDataNorm(:,i) = cellfun(@(x) x./thisMax, thisData,'un',0);
+end
+
+postNorm = thisDataNorm;
+postHeaders = cellfun(@(x) sprintf('%s_Norm',x),thisCell,'un',0)';
+
+%%
+
+xlswrite(fname,antHeaders,'antNorm');
+xlswrite(fname,postHeaders,'postNorm');
+xlswrite(fname,antNorm,'antNorm','A2');
+xlswrite(fname,postNorm,'postNorm','A2');