%% 3AOB JFC clear all; clc addpath('Z:\EXPERIMENTS\mTBICoBRE\EEG\'); addpath(genpath('Y:\Programs\eeglab12_0_2_1b')); rmpath('Y:\Programs\eeglab12_0_2_1b\functions\octavefunc'); rmpath('Y:\Programs\eeglab14_0_0b\functions\octavefunc'); datadir='Y:\EEG_Data\mTBICoBRE\'; % Data are here saveloc='Z:\EXPERIMENTS\mTBICoBRE\EEG\3AOB Preproc\'; load('Z:\EXPERIMENTS\mTBICoBRE\EEG\BV_Chanlocs_60.mat'); cd(saveloc); sx_dirs=dir([datadir,'M*']); for sxi=1:length(sx_dirs) for ses=1:3 sessdir=[datadir,sx_dirs(sxi).name,'\eeg\RawEEG\']; sx_sess{sxi}{1,ses}=dir([sessdir,'*_',num2str(ses),'_ODDBALL.vhdr']); sx_sess{sxi}{2,ses}=sessdir; sx_sess{sxi}{3,ses}=sx_dirs(sxi).name; end end LOG=[]; for sxi=1:size(sx_sess,2) for sess=1:3 if ~isempty( sx_sess{sxi}{1,sess} ) subno=str2num(sx_sess{sxi}{1,sess}.name(1:4)); URSI=sx_sess{sxi}{3,sess}; LOG(subno-3000,sess+1)=subno; LOG(subno-3000,1)=str2num(URSI(end-4:end)); end end end for sxi=1:size(sx_sess,2) for sess=1:3 if ~isempty( sx_sess{sxi}{1,sess} ) subno=str2num(sx_sess{sxi}{1,sess}.name(1:4)); thisdir=sx_sess{sxi}{2,sess}; URSI=sx_sess{sxi}{3,sess}; LOG2(sxi,sess)=subno; if ~exist([saveloc,num2str(subno),'_',num2str(sess),'_3AOB.mat']); % Data are 65 chans: 1=63 is EEG, 64 is VEOG, 65 is EKG Ref'd to CPz - - will want to retrieve that during re-referencing EEG = pop_loadbv(thisdir,[num2str(subno),'_',num2str(sess),'_ODDBALL.vhdr']); clc; disp(['Loading ',num2str(subno),' s',num2str(sess)]); % Run PATCH for sx<3003 s<2 AND for bad templates PATCH % Get Locs locpath=('Y:\Programs\eeglab12_0_2_1b\plugins\dipfit2.2\standard_BESA\standard-10-5-cap385.elp'); EEG = pop_chanedit(EEG, 'lookup', locpath); EEG = eeg_checkset( EEG ); % Get event types for ai=2:length(EEG.event); clear temp; temp=EEG.event(ai).type; if isempty(strmatch('boundary',temp)); TYPES(ai)=str2num(temp(2:end)) ; clear temp; end end UNIQUE_TYPES=unique(TYPES); for ai=1:length(UNIQUE_TYPES); UNIQUE_TYPES_COUNT(ai)=sum(TYPES==UNIQUE_TYPES(ai)); end clc; TRIGGERS=[UNIQUE_TYPES;UNIQUE_TYPES_COUNT] % Trigger type, Frequency % Epoch All_STIM={'S201','S200','S202'}; % Std, Target, Novel EEG = pop_epoch( EEG, All_STIM, [-2 2], 'newname', 'Epochs', 'epochinfo', 'yes'); EEG = eeg_checkset( EEG ); % Remove VEOG and EKG EEG.EKG=squeeze(EEG.data(65,:,:)); EEG.VEOG=squeeze(EEG.data(64,:,:)); EEG.data=EEG.data(1:63,:,:); EEG.nbchan=63; EEG.chanlocs(65)=[]; EEG.chanlocs(64)=[]; % Fix BV-specific issue - - - only needed for APPLE for ai=1:size(EEG.urevent,2), EEG.urevent(ai).bvtime=EEG.urevent(ai).bvmknum; end for ai=1:size(EEG.event,2), EEG.event(ai).bvtime=EEG.event(ai).bvmknum; end for ai=1:size(EEG.epoch,2), EEG.epoch(ai).eventbvtime=EEG.epoch(ai).eventbvmknum; end % Add CPz EEG = pop_chanedit(EEG,'append',63,'changefield',{64 'labels' 'CPz'}); EEG = pop_chanedit(EEG,'lookup', locpath); % Re-Ref to Average Ref and recover CPz EEG = pop_reref(EEG,[],'refloc',struct('labels',{'CPz'},'type',{''},'theta',{180},'radius',{0.12662},'X',{-32.9279},'Y',{-4.0325e-15},'Z',{78.363},... 'sph_theta',{-180},'sph_phi',{67.208},'sph_radius',{85},'urchan',{64},'ref',{''}),'keepref','on'); % Remove everything else NOW that CPz has been reconstructed from the total EEG.MASTOIDS = squeeze(mean(EEG.data([10,21],:,:),1)); EEG.data = EEG.data([1:4,6:9,11:20,22:26,28:64],:,:); EEG.nbchan=60; EEG.chanlocs(27)=[]; EEG.chanlocs(21)=[]; EEG.chanlocs(10)=[]; EEG.chanlocs(5)=[]; % Have to be in this order! % Should probably re-ref to average again now that the contaminated channels are gone EEG = pop_reref(EEG,[]); % Remove mean EEG = pop_rmbase(EEG,[],[]); % ---------------------- % Setup APPLE to interp chans, reject epochs, & ID bad ICs. Output will be Avg ref'd and ICA'd. eeg_chans=1:60; Do_ICA=1; ref_chan=36; % Re-Ref to FCz [WEIRD STEP, BUT THIS IS FOR FASTER, which is a part of APPLE] EEG = pop_reref(EEG,ref_chan,'keepref','on'); % Run APPLE (will re-ref data to avg ref) [EEG,EEG.bad_chans,EEG.bad_epochs,EEG.bad_ICAs]=APPLE_3AOB(EEG,eeg_chans,ref_chan,Do_ICA,subno,EEG.VEOG,sess,BV_Chanlocs_60); % Save save([num2str(subno),'_',num2str(sess),'_3AOB.mat'],'EEG'); % ---------------------- %% Remove the (presumptive) bad ICAs: bad_ICAs_To_Remove=EEG.bad_ICAs{2}; if bad_ICAs_To_Remove==0, bad_ICAs_To_Remove=1; end EEG = pop_subcomp( EEG, bad_ICAs_To_Remove, 0); % Get the good info out of the epochs for ai=1:size(EEG.epoch,2) % Initialize EEG.epoch(ai).CUE=NaN; for bi=1:size(EEG.epoch(ai).eventlatency,2) % Get STIMTYPE if EEG.epoch(ai).eventlatency{bi}==0 && isempty(strmatch(EEG.epoch(ai).eventtype{bi},'N999')); % If this bi is the event % Get StimType FullName=EEG.epoch(ai).eventtype{bi}; EEG.epoch(ai).CUE=str2num(FullName(2:end)) ; clear FullName VECTOR(ai,1)=EEG.epoch(ai).CUE; end end end % Let's just do this for display dims=size(EEG.data); EEG.data=eegfilt(EEG.data,500,[],20); EEG.data=reshape(EEG.data,dims(1),dims(2),dims(3)); % Set Params tx=-2000:2:1998; b1=find(tx==-200); b2=find(tx==0); t1=find(tx==-500); t2=find(tx==1000); toporange1=find(tx==250); toporange2=find(tx==600); toporangetot=250:2:600; tx2disp=-500:2:1000; MAPLIMS=[-8 8]; % Basecor your ERPs here so they are pretty. BASE=squeeze( mean(EEG.data(:,b1:b2,:),2) ); for ai=1:dims(1) EEG.data(ai,:,:)=squeeze(EEG.data(ai,:,:))-repmat( BASE(ai,:),dims(2),1 ); end % Get max of P2 across all condis site=11; % Pz ERP4topo=mean(EEG.data(site,toporange1:toporange2,VECTOR(:,1)==200),3); topomax_P3b=toporangetot(find(ERP4topo==max(ERP4topo))); topotoplot_P3b=find(tx==topomax_P3b); site=36; % FCz ERP4topo=mean(EEG.data(site,toporange1:toporange2,VECTOR(:,1)==202),3); topomax_P3a=toporangetot(find(ERP4topo==max(ERP4topo))); topotoplot_P3a=find(tx==topomax_P3a); % -------------- figure; site=11; % Pz subplot(3,4,1:4); hold on plot(tx2disp,mean(EEG.data(site,t1:t2,VECTOR(:,1)==201),3),'k'); plot(tx2disp,mean(EEG.data(site,t1:t2,VECTOR(:,1)==200),3),'r'); plot(tx2disp,mean(EEG.data(site,t1:t2,VECTOR(:,1)==202),3),'b'); plot([topomax_P3b topomax_P3b],[-2 2],'m','linewidth',2); % indicate the max with a magenta line title(['Pz Subno: ',num2str(subno),' Sess:',num2str(sess)]); legend({'Std','Target','Novel'},'Location','NorthWest'); % -------------- site=36; % FCz subplot(3,4,5:8); hold on plot(tx2disp,mean(EEG.data(site,t1:t2,VECTOR(:,1)==201),3),'k'); plot(tx2disp,mean(EEG.data(site,t1:t2,VECTOR(:,1)==200),3),'r'); plot(tx2disp,mean(EEG.data(site,t1:t2,VECTOR(:,1)==202),3),'b'); plot([topomax_P3a topomax_P3a],[-2 2],'m','linewidth',2); % indicate the max with a magenta line title(['FCz Subno: ',num2str(subno),' Sess:',num2str(sess)]); % -------------- subplot(3,4,9); hold on topoplot( mean(EEG.data(:,topotoplot_P3b,VECTOR(:,1)==201),3) , BV_Chanlocs_60,'maplimits',MAPLIMS); title('Std @ P3b') subplot(3,4,10); hold on topoplot( mean(EEG.data(:,topotoplot_P3b,VECTOR(:,1)==200),3) , BV_Chanlocs_60,'maplimits',MAPLIMS); title('Targ') subplot(3,4,11); hold on topoplot( mean(EEG.data(:,topotoplot_P3a,VECTOR(:,1)==202),3) , BV_Chanlocs_60,'maplimits',MAPLIMS); title('Novel') saveas(gcf, [num2str(subno),'_',num2str(sess),'_3AOB_ERPs.png'],'png'); close all; clear EEG VECTOR BASE PROBE TRIGGERS TYPES UNIQUE* did* topo* ERP* URSI dims eeg_chans; end end end end %%