gk_getTunedROIsDirect.m

function xpr = gk_getTunedROIsDirect(sig,iscell,stim,before,after,p_thr)
% USAGE: xpr = gk_getTunedROIsDirect(sig,iscell,stim,before,after,p_thr)
%
% Uses anova1 on the responses and if significant based on p_thr assigns
% the ROI as tuned
%
% INPUT:
%   sig :       loaded from SIG.npy (SIG: F, Fneu, spks)
%   iscell:     loaded from iscell.npy
%   before:     seconds before stim onset
%   after:      seconds after stim offset
%   pthr:       p-value to select ROIs as tuned

% Author: Georgios A. Keliris

sig = sig(logical(iscell(:,1)),:);
xpr = gk_getSigAllTrials(sig,stim,before,after);
xpr.cellIDs=find(iscell(:,1));

if isfield(xpr,'stimAngles')
    nGrps=numel(unique(xpr.stimAngles));
else
    nGrps=1;
end

for g=1:nGrps
    
    for r=1:size(xpr.cellIDs,1)
        roi=gk_getCellTrials(xpr,r);
        grp=roi.responses(roi.responses.grpID==g,1:2);
        pON(r,g)=anova1(grp.ONresp, grp.stimID,"off");
    end
    xpr.grp(g).pON=pON(:,g);
    xpr.grp(g).isOnTuned=xpr.grp(g).pON<p_thr;
    xpr.grp(g).onTunedIDs = find(xpr.grp(g).isOnTuned);
    xpr.grp(g).pThr=p_thr;
    [~, xpr.grp(g).sortedOnTunedIDs]= sort(xpr.grp(g).pON(xpr.grp(g).onTunedIDs));
end
xpr.pON=pON;
xpr.pONmin=min(pON,[],2);
xpr.pThr=p_thr;
xpr.isOnTuned_allGrp=xpr.pONmin<p_thr;
xpr.onTunedIDs_allGrp = find(xpr.isOnTuned_allGrp);
[~, xpr.sortedOnTunedIDs_allGrp]= sort(xpr.pONmin(xpr.onTunedIDs_allGrp));
xpr.tunedGlobalIDs=xpr.cellIDs(xpr.onTunedIDs_allGrp(xpr.sortedOnTunedIDs_allGrp));

end