/
beta_extraction_peak_ExpReg.m
214 lines (170 loc) · 6.99 KB
/
beta_extraction_peak_ExpReg.m
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
%% Extract beta estimates for experimental regressors
clear all
close all
clc
%% Data info
SJs = { 'S01' 'S02' 'S03' 'S04' 'S05' 'S06' 'S08' 'S09' 'S11' 'S13' 'S14' 'S15' 'S16' 'S17' 'S18' 'S19' 'S20' 'S21' 'S22' 'S23' 'S24' 'S25' 'S26' 'S28' 'S29' 'S30' 'S32'};
bms_name = 'BMS_FFX_IntDetPFUncRep_RT_2x2x2';
data_dir = 'my_data_dir'; % Data directory
roi_dir = 'my_roi_dir'; % Directory containnig masks of ROIs
model_dir = 'BayesGLM'; % Directory containing first level beta images
map_dir = fullfile(model_dir, 'BMS', bms_name); % Directory containing single subject BMS results (PPMs)
trg_dir = fullfile(data_dir, 'ExpReg betas');
if ~exist(trg_dir,'dir')
mkdir(trg_dir)
end
name = 'BayesGLM_Betas_BMSPeaks_ExpRegs';
info = 'Beta estimates of experimental regressors from individual BMS peaks within group level BMS ROIs';
% Bayesian 1st level models to extract parameters from
models = {'Int'
'PF'
'Det'
'Unc'
'Rep'};
mod = cellfun(@strcat,repmat({'mpctd_'},numel(models),1),models,repmat({'RT'},numel(models),1),'UniformOutput',0); % Complete file names
% ROI name model
ROIs = { 'Int_rSI' 'R SIa' 1
'Int_rSIIp' 'R SIIp' 1
'Int_rSIIa' 'R SIIa' 1
'Int_lSII' 'L SIIm' 1
'PF_rSI' 'R SIp' 2
'PF_rSII' 'R SII' 2
'PF_lSII' 'L SII' 2
'Det_rSIIs' 'R SIIs' 3
'Det_rSIIi' 'R SIIi' 3
'Det_lSII' 'L SIIl' 3
'Det_lMFG' 'L SFG' 3
'Det_lIPS' 'L IPL' 3
'Det_lLG' 'L V3' 3
'Unc_MSF' 'SMG/ACC' 4
'Unc_lAIC' 'L AIC' 4
'Unc_rAIC' 'R AIC' 4
'Rep_lSMA' 'L SMA' 5
'Rep_lThal' 'L Thal' 5
'Rep_rSMG' 'R SMaG' 5 };
nROI = size(ROIs,1);
nSub = numel(SJs);
nRuns = 4;
% beta images
prefix_Cbeta = 'srCbeta_';
prefix_SDbeta = 'srSDbeta_';
b = 3; % beta idx expreg
%% Get estimates
if exist(fullfile(trg_dir,[name '.mat']),'file')
load(fullfile(trg_dir,[name '.mat']))
disp(['Loaded ' name 'betas!'])
else
P = cell2struct(cell(size(ROIs,1),1),ROIs(:,1),1);
P.info = info;
for r = 1:nROI
roi = ROIs{r,1};
m = ROIs{r,3};
model = ['BayesFFX' mod{m}];
map = mod{m};
disp(roi)
% Get peaks
ROImax = find_bms_peaks(map, map_dir, roi, roi_dir, SJs);
coords = ROImax.coords';
peaks = cell2mat(coords(:,2))';
% Prepare data structure
P.(ROIs{r,1}).beta = nan(nSub, nRuns);
P.(ROIs{r,1}).sd = nan(nSub, nRuns);
P.(ROIs{r,1}).roi = ROIs{r,1};
% Assemble beta estimates for BMS peaks in ROIs per subject and run
for s = 1:nSub
disp(SJs{s})
% get betas
sub_dir = fullfile(data_dir, SJs{s}, model_dir, model);
for r = 1:nRuns
beta_idx = b+(r-1)*22;
Cbeta_img = spm_select('FPList', sub_dir, ['^' prefix_Cbeta '0{1,3}' num2str(beta_idx) '.nii$']);
SDbeta_img = spm_select('FPList', sub_dir, ['^' prefix_SDbeta '0{1,3}' num2str(beta_idx) '.nii$']);
P.(ROIs{r,1}).beta(s,r) = spm_summarise(Cbeta_img, peaks(:,s));
P.(ROIs{r,1}).sd(s,r) = spm_summarise(SDbeta_img, peaks(:,s));
end
end
% Aggregate runs
P.(roi).beta_r = squeeze(mean(P.(roi).beta,2))';
P.(roi).sd_r = squeeze(mean(P.(roi).sd,2))';
% Aggregate subjects
P.(roi).beta_rs = mean(P.(roi).beta_r,1);
P.(roi).sd_rs = mean(P.(roi).sd_r,1);
end
end
save(fullfile(trg_dir,[name '.mat']),'P');
%% Perform Bayesian t-tests on betas
bf10 = nan(nROI,1);
pValue = nan(nROI,1);
fprintf('\n\n Bayes factors and p-values \n\n')
for r = 1:nROI
[bf10(r),pValue(r)] = bf_ttest(P.(ROIs{r,1}).beta_r);
bf01 = 1./bf10;
fprintf('%s mean beta = %.3f BF10 = %.3f BF01 = %.3f p = %.3f\n',ROI_names{r},mean(P.(ROIs{r,1}).beta_r),bf10(r),bf01(r),pValue(r));
end
%% Scatter plot per ROI
betas = nan(nSub,nROI);
for r = 1:nROI
betas(:,r) = P.(ROIs{r,1}).beta_r;
end
cols = {[0 1 0] % Green: Intensity
[0 0 1] % Blue: P(Detection)
[1 0 0] % Red: Detection
[0 1 1] % Cyan: Uncertainty
[1 0 1]}; % Magenta: Report
for r = 1:nROI
f = figure;
hold on
set(gca,'FontSize',8,'FontName','Calibri')
% Scatter plot
scatter(ones(nSub,1),betas(:,r),5, 'MarkerFaceColor','none','MarkerEdgeColor',cols{ROIs{r,3},:},'LineWidth',1.5)
scatter(1,mean(betas(:,r)),5, 'MarkerFaceColor',[0 0 0],'MarkerEdgeColor',[0 0 0],'LineWidth',1.5)
% Axes and Ticks
xlim([.999999 1.000001])
line(xlim,[0 0],'LineWidth',2/3,'Color','k')
min_b = min(betas(:,r));
max_b = max(betas(:,r));
range_b = max_b - min_b;
margin = range_b/20;
bf_topup = 3*margin;
ylim([(min_b - margin) (max_b + margin + bf_topup )])
ylim_min = (ceil(min_b*10)/10);
ylim_max = (floor(max_b*10)/10);
if ylim_min < 0 && ylim_max > 0
ylabs = [ylim_min 0 ylim_max];
else
ylabs = [ylim_min ylim_max];
end
set(gca,'YTick',ylabs)
set(gca,'XTick',[])
ylabs_str = cellstr(num2str(ylabs'));
ylabs_str = cellfun(@strrep,ylabs_str, repmat({'0.'},numel(ylabs_str),1), repmat({'.'},numel(ylabs_str),1),'UniformOutput',0);
set(gca,'YTickLabel',{})
xlims = xlim;
text(repmat(xlims(1)-0.0000002,numel(ylabs_str),1),ylabs,ylabs_str,'HorizontalAlignment','right','FontName','Calibri','FontSize',8)
set(gca,'xcolor',[1 1 1])
% Figure size
set(gcf,'Units','centimeters');
x0 = 1;
y0 = 1;
width = .4;
height = 2;
set(gca,'Units','centimeters','position', [x0,y0,width,height])
set(gcf,'Units','centimeters','position',[0,0,10.5,8])
title(ROI_names{r})
saveas(f,fullfile(trg_dir, 'model betas', [name,'_',ROIs{r,1},'.emf']))
end
%% Test unimodality and normality
nboot = 1000;
dip = nan(nROI,1);
p_dip = nan(nROI,1);
lillie = nan(nROI,1);
p_lillie = nan(nROI,1);
for r = 1:nROI
% Unimodality
[dip(r), p_dip(r)] = HartigansDipSignifTest(betas(:,r), nboot);
subplot(4,5,r)
hist(betas(:,r))
title([ROIs{r,2} ': dip=',num2str(dip(r),3), ', p=',num2str(p_dip(r),3)])
% Normality
[lillie(r), p_lillie(r)] = lillietest(betas(:,r));
end