-
Notifications
You must be signed in to change notification settings - Fork 27
/
limo_stat_values.m
489 lines (455 loc) · 21 KB
/
limo_stat_values.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
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
function [M, mask, mytitle] = limo_stat_values(varargin)
% find corrected p values and mask from data under H0
%
% FORMAT [M, mask, mytitle] = limo_stat_values(FileName,p,MCC,LIMO)
%
% INPUTS
% FileName = Name of the file selected
% p = p value for thresholding
% MCC = multiple comparisons option
% 1 none
% 2 clustering
% 3 TFCE
% 4 Max
% LIMO = LIMO.mat structure
%
% OUTPUTS
% M = the (corrected) p values
% mask = the significant data
% mytitle = title that include the filename, effect and method used
%
% see limo_display_results
%
% Cyril Pernet, Andrew Stewart, Marianne Latinus, Guilaume Rousselet
% ------------------------------------------------------------------
% Copyright (C) LIMO Team 2020
root = fileparts(which('limo_eeg'));
pathCell = regexp(path, pathsep, 'split');
onPath = all([sum(strcmp([root filesep 'help'],pathCell))~=0,...
sum(strcmp([root filesep 'limo_cluster_functions'],pathCell))~=0,...
sum(strcmp([root filesep 'external' filesep 'psom'],pathCell))~=0]);
if onPath == 0
addpath([root filesep 'limo_cluster_functions'])
addpath([root filesep 'external'])
addpath([root filesep 'external' filesep 'psom'])
addpath([root filesep 'help'])
end
FileName = varargin{1}; % Name of the file selected
p = varargin{2}; % p value
MCC = varargin{3}; % multiple comparison option
LIMO = varargin{4}; % LIMO.mat
plotFlag = true; % always display if clustering fails
% check the appropriate method is used
% -----------------------------------
if isfield(LIMO,'Type')
if strcmp(LIMO.Type,'Components') && MCC ~= 1
MCC = 4; % for ICA only max stat since we can't cluster them based a topography
disp('only maximum statistics can be used for ICA')
end
end
% check that a neighbouring matrix is there for clustering
% -------------------------------------------------------
if MCC == 2
limo_check_neighbourghs(LIMO)
end
% load data and set outputs to empty
% ----------------------------------
if exist(FileName,'file')
matfile = load(FileName);
else
matfile = load(fullfile(LIMO.dir,FileName));
end
M = [];
mask = [];
mytitle = [];
disp(' ');
% disp some references for this
% -----------------------------
if MCC ~= 1
if MCC == 2
disp('Ref for Clustering & Bootstrap:')
disp('Maris, E. & Oostenveld, R. 2007')
disp('Nonparametric statistical testing of EEG- and MEG-data.')
disp('Journal of Neuroscience Methods, 164, 177-190')
disp(' ');
disp('Pernet, C., Latinus, M., Nichols, T. & Rousselet, G.A. (2015).')
disp('Cluster-based computational methods for mass univariate analyses')
disp('of event-related brain potentials/fields: A simulation study.')
disp('Journal of Neuroscience methods, 250, 83-95')
disp(' ');
elseif MCC == 3
disp('Ref for TFCE:')
disp('Pernet, C., Latinus, M., Nichols, T. & Rousselet, G.A. (2015).')
disp('Cluster-based computational methods for mass univariate analyses')
disp('of event-related brain potentials/fields: A simulation study.')
disp('Journal of Neuroscience methods, 250, 83-95')
disp(' ');
end
end
if MCC ~= 1
fprintf('computing corrected statistics at %s...\n',datetime('now','Format','hh:mm:ss'));
end
%% Deal with each case of FileName
% -------------------------------
%% GLM (from 1st or 2nd level) also robust regresion, robust ANOVA
% ------------------------------------------------------------------------
if strcmpi(LIMO.Analysis,'Time-Frequency')
if strcmp(FileName,'R2.mat')
M = squeeze(matfile.R2(:,:,:,2)); % F values
Pval = squeeze(matfile.R2(:,:,:,3)); % P values
MCC_data = 'H0_R2.mat';
titlename = 'R^2 Coef';
elseif strncmp(FileName,'Condition_effect',16)
effect_nb = eval(FileName(18:end-4));
M = squeeze(matfile.Condition_effect(:,:,:,1));
Pval = squeeze(matfile.Condition_effect(:,:,:,2));
MCC_data = sprintf('H0_Condition_effect_%g.mat',effect_nb);
titlename = sprintf('Condition effect %g F values',effect_nb);
elseif strncmp(FileName,'Covariate_effect',16)
effect_nb = eval(FileName(18:end-4));
M = squeeze(matfile.Covariate_effect(:,:,:,1));
Pval = squeeze(matfile.Covariate_effect(:,:,:,2));
MCC_data = sprintf('H0_Covariate_effect_%g.mat',effect_nb);
titlename = sprintf('Covariate effect %g F values',effect_nb);
elseif strncmp(FileName,'Interaction_effect',18)
effect_nb = eval(FileName(20:end-4));
M = squeeze(matfile.Interaction_effect(:,:,:,1));
Pval = squeeze(matfile.Interaction_effect(:,:,:,2));
MCC_data = sprintf('H0_Interaction_effect_%g.mat',effect_nb);
titlename = sprintf('Interaction effect %g F values',effect_nb);
elseif strncmp(FileName,'semi_partial_coef',17)
effect_nb = eval(FileName(19:end-4));
M = squeeze(matfile.semi_partial_coef(:,:,:,2));
Pval = squeeze(matfile.semi_partial_coef(:,:,:,3));
MCC_data = sprintf('H0_semi_partial_coef_%g.mat',effect_nb);
titlename = sprintf('Semi Partial Coef %g',effect_nb);
elseif strncmp(FileName,'con_',4)
effect_nb = eval(FileName(5:end-4));
M = squeeze(matfile.con(:,:,:,4));
Pval = squeeze(matfile.con(:,:,:,5));
MCC_data = sprintf('H0_con_%g.mat',effect_nb);
titlename = sprintf('Contrast %g T values',effect_nb);
elseif contains(FileName,'ttest') || contains(FileName,'LI_Map')
matfile = matfile.(cell2mat(fieldnames(matfile)));
M = matfile(:,:,:,4); % T values
Pval = matfile(:,:,:,5);
MCC_data = sprintf('H0_%s', FileName);
name = FileName(1:strfind(FileName,'ttest')+4);
name(strfind(name,'_')) = ' ';
titlename = sprintf('%s t-test T values',name);
elseif strncmp(FileName,'ess_',4)
effect_nb = eval(FileName(5:end-4));
M = squeeze(matfile.ess(:,:,:,end-1));
Pval = squeeze(matfile.ess(:,:,:,end));
MCC_data = sprintf('H0_ess_%g.mat',effect_nb);
titlename = sprintf('Contrast %g F values',effect_nb);
end
else % same with one less dimention
if strcmp(FileName,'R2.mat')
M = squeeze(matfile.R2(:,:,2)); % F values
Pval = squeeze(matfile.R2(:,:,3)); % P values
MCC_data = 'H0_R2.mat';
titlename = 'R^2 Coef';
elseif strncmp(FileName,'Condition_effect',16)
effect_nb = eval(FileName(18:end-4));
M = squeeze(matfile.Condition_effect(:,:,1));
Pval = squeeze(matfile.Condition_effect(:,:,2));
MCC_data = sprintf('H0_Condition_effect_%g.mat',effect_nb);
titlename = sprintf('Condition effect %g F values',effect_nb);
elseif strncmp(FileName,'Covariate_effect',16)
effect_nb = eval(FileName(18:end-4));
M = squeeze(matfile.Covariate_effect(:,:,1));
Pval = squeeze(matfile.Covariate_effect(:,:,2));
MCC_data = sprintf('H0_Covariate_effect_%g.mat',effect_nb);
titlename = sprintf('Covariate effect %g F values',effect_nb);
elseif strncmp(FileName,'Interaction_effect',18)
effect_nb = eval(FileName(20:end-4));
M = squeeze(matfile.Interaction_effect(:,:,1));
Pval = squeeze(matfile.Interaction_effect(:,:,2));
MCC_data = sprintf('H0_Interaction_effect_%g.mat',effect_nb);
titlename = sprintf('Interaction effect %g F values',effect_nb);
elseif strncmp(FileName,'semi_partial_coef',17)
effect_nb = eval(FileName(19:end-4));
M = squeeze(matfile.semi_partial_coef(:,:,2));
Pval = squeeze(matfile.semi_partial_coef(:,:,3));
MCC_data = sprintf('H0_semi_partial_coef_%g.mat',effect_nb);
titlename = sprintf('Semi Partial Coef %g',effect_nb);
elseif strncmp(FileName,'con_',4)
effect_nb = eval(FileName(5:end-4));
M = squeeze(matfile.con(:,:,4));
Pval = squeeze(matfile.con(:,:,5));
MCC_data = sprintf('H0_con_%g.mat',effect_nb);
titlename = sprintf('Contrast %g T values',effect_nb);
elseif contains(FileName,'ttest') || contains(FileName,'LI_Map')
matfile = matfile.(cell2mat(fieldnames(matfile)));
M = matfile(:,:,4); % T values
Pval = matfile(:,:,5);
MCC_data = sprintf('H0_%s', FileName);
name = FileName(1:strfind(FileName,'ttest')+4);
name(strfind(name,'_')) = ' ';
titlename = sprintf('%s T values',name);
elseif strncmp(FileName,'ess_',4)
effect_nb = eval(FileName(max(strfind(FileName,'_'))+1:end-4));
M = squeeze(matfile.ess(:,:,end-1));
Pval = squeeze(matfile.ess(:,:,end));
MCC_data = sprintf('H0_ess_%g.mat',effect_nb);
titlename = sprintf('Contrast %g F values',effect_nb);
end
end
% no correction for multiple testing
% -----------------------------------
if ~isempty(M) && MCC == 1
M = Pval;
mask = Pval <= p;
mytitle = sprintf('%s\nuncorrected threshold',titlename);
% cluster correction for multiple testing
% ---------------------------------------
elseif ~isempty(M) && MCC == 2
if exist(['H0' filesep MCC_data],'file')
try
H0_data = load(['H0' filesep MCC_data]);
H0_data = H0_data.(cell2mat(fieldnames(H0_data)));
if strcmpi(LIMO.Analysis,'Time-Frequency')
if contains(FileName,'R2') || contains(FileName,'semi_partial')
bootM = squeeze(H0_data(:,:,:,2,:)); % get all F values under H0
bootP = squeeze(H0_data(:,:,:,3,:)); % get all P values under H0
else
bootM = squeeze(H0_data(:,:,:,1,:));
bootP = squeeze(H0_data(:,:,:,2,:));
end
if size(M,1) == 1
tmp = NaN(1,size(M,2),size(M,3),size(bootM,3));
tmp(1,:,:,:) = bootM; bootM = tmp;
tmp(1,:,:,:) = bootP; bootP = tmp;
clear tmp
end
else
if contains(FileName,'R2') || contains(FileName,'semi_partial')
bootM = squeeze(H0_data(:,:,2,:)); % get all F values under H0
bootP = squeeze(H0_data(:,:,3,:)); % get all P values under H0
else
bootM = squeeze(H0_data(:,:,1,:));
bootP = squeeze(H0_data(:,:,2,:));
end
if size(M,1) == 1
tmp = NaN(1,size(M,2),size(bootM,2));
tmp(1,:,:,:) = bootM; bootM = tmp;
tmp(1,:,:,:) = bootP; bootP = tmp;
clear tmp
end
end
% finally get cluster mask and corrected p-values
if contains(FileName,'ttest') || contains(FileName,'LI_Map')
[mask,M] = limo_clustering(M.^2,Pval,bootM.^2,bootP,LIMO,MCC,p,plotFlag); % mask and cluster p values
else
[mask,M] = limo_clustering(M,Pval,bootM,bootP,LIMO,MCC,p,plotFlag); % mask and cluster p values
end
Nclust = unique(mask);
Nclust = length(Nclust)-1; % mask = mask>0;
if Nclust <= 1
Mclust = 'cluster';
else
Mclust = 'clusters';
end
mytitle = sprintf('%s\ncluster correction (%g %s)', titlename, Nclust, Mclust);
catch ME
limo_errordlg(sprintf('error log: %s \n',ME.message),'cluster correction failure')
return
end
else
limo_errordlg(['H0' filesep MCC_data ' not found'],'cluster correction failure')
return
end
% correction using the max
% --------------------------
elseif ~isempty(M) && MCC == 4 % Stat max
if exist(['H0' filesep MCC_data],'file')
try
H0_data = load(['H0' filesep MCC_data]);
H0_data = H0_data.(cell2mat(fieldnames(H0_data)));
if strcmpi(LIMO.Analysis,'Time-Frequency')
if contains(FileName,'R2') || contains(FileName,'semi_partial')
bootM = squeeze(H0_data(:,:,:,2,:)); % get all F values under H0
else
bootM = squeeze(H0_data(:,:,:,1,:));
end
else
if contains(FileName,'R2') || contains(FileName,'semi_partial')
bootM = squeeze(H0_data(:,:,2,:)); % get all F values under H0
else
bootM = squeeze(H0_data(:,:,1,:));
end
end
clear H0_data;
[mask,M] = limo_max_correction(abs(M),abs(bootM),p,plotFlag);
mytitle = sprintf('%s\ncorrection by max',titlename);
catch ME
limo_errordlg(sprintf('error log: %s \n',ME.message),'max correction failure')
return
end
else
limo_errordlg(['H0' filesep MCC_data ' not found'],'max correction failure')
end
% correction using TFCE
% --------------------------
elseif ~isempty(M) && MCC == 3 % Stat max
if exist(fullfile(LIMO.dir,['tfce' filesep 'tfce_' FileName]),'file')
try
score = load(fullfile(LIMO.dir,['tfce' filesep 'tfce_' FileName]));
score = score.(cell2mat(fieldnames(score)));
H0_score = load(fullfile(LIMO.dir,['H0' filesep 'tfce_H0_' FileName]));
H0_score = H0_score.(cell2mat(fieldnames(H0_score)));
[mask,M] = limo_max_correction(score,H0_score,p,plotFlag);
mytitle = sprintf('%s\ncorrection using TFCE',titlename);
catch ME
limo_errordlg(sprintf('error log: %s \n',ME.message),'tfce correction failure')
return
end
else
limo_errordlg('no tfce tfce file was found','missing data')
end
end
% ------------------------
%% Repeated measures ANOVA
% ------------------------
if contains(FileName,'Rep_ANOVA')
% all files have dim electrode x [freq/time] frames x F/p
if strcmp(LIMO.Analysis,'Time-Frequency') || strcmp(LIMO.Analysis,'ITC')
M = matfile.(cell2mat(fieldnames(matfile)))(:,:,:,1);
PVAL = matfile.(cell2mat(fieldnames(matfile)))(:,:,:,2);
else
M = matfile.(cell2mat(fieldnames(matfile)))(:,:,1);
PVAL = matfile.(cell2mat(fieldnames(matfile)))(:,:,2);
end
MCC_data = fullfile(LIMO.dir,['H0' filesep 'H0_' FileName]);
% no correction for multiple testing
% -----------------------------------
if MCC == 1
mask = PVAL <= p;
M = PVAL;
if contains(FileName,'Rep_ANOVA_Interaction')
mytitle = sprintf('Interaction F-values\nuncorrected threshold');
elseif contains(FileName,'Rep_ANOVA_Gp_effect')
mytitle = sprintf('Gp effect F-values\nuncorrected threshold');
elseif contains(FileName,'Rep_ANOVA_Main')
mytitle = sprintf('Main Effect F-values\nuncorrected threshold');
end
% cluster correction for multiple testing
% ---------------------------------------
elseif MCC == 2
if exist(MCC_data,'file')
try
H0_data = load(MCC_data);
H0_data = H0_data.(cell2mat(fieldnames(H0_data)));
if strcmpi(LIMO.Analysis,'Time-Frequency')
bootT = squeeze(H0_data(:,:,:,1,:));
bootP = squeeze(H0_data(:,:,:,2,:));
if size(M,1) == 1
tmp = NaN(1,size(M,2),size(M,3),size(bootT,4));
tmp(1,:,:,:) = bootT; bootT = tmp;
tmp(1,:,:,:) = bootP; bootP = tmp;
clear tmp
end
else
bootT = squeeze(H0_data(:,:,1,:));
bootP = squeeze(H0_data(:,:,2,:));
if size(M,1) == 1
tmp = NaN(1,size(M,2),size(bootT,2));
tmp(1,:,:) = bootT; bootT = tmp;
tmp(1,:,:) = bootP; bootP = tmp;
clear tmp
end
end
if size(M,1) == 1
[mask,M] = limo_clustering(M,PVAL,bootT,bootP,LIMO,3,p,plotFlag); % temporal clustering
else
[mask,M] = limo_clustering(M,PVAL,bootT,bootP,LIMO,2,p,plotFlag); % spatial-temporal clustering
end
Nclust = unique(mask); Nclust = length(Nclust)-1; % mask = mask>0;
if Nclust <= 1; Mclust = 'cluster'; else ; Mclust = 'clusters'; end
if contains(FileName,'Rep_ANOVA_Interaction')
mytitle = sprintf('Interaction F-values\ncluster correction (%g %s)', Nclust, Mclust);
elseif contains(FileName,'Rep_ANOVA_Gp_effect')
mytitle = sprintf('Gp effect F-values\ncluster correction (%g %s)', Nclust, Mclust);
elseif contains(FileName,'Rep_ANOVA_Main')
mytitle = sprintf('Main effect F-values\ncluster correction (%g %s)', Nclust, Mclust);
end
catch ME
limo_errordlg(sprintf('error log: %s \n',ME.message),'cluster correction failure')
return
end
else
limo_errordlg(['H0' filesep MCC_data ' not found'],'cluster correction failure')
return
end
% T max correction for multiple testing
% -------------------------------------
elseif MCC == 4 % Stat max
if exist(MCC_data,'file')
try
H0_data = load(MCC_data);
H0_data = H0_data.(cell2mat(fieldnames(H0_data)));
if strcmpi(LIMO.Analysis,'Time-Frequency')
bootT = squeeze(H0_data(:,:,:,1,:));
if size(M,1) == 1
tmp = NaN(1,size(M,2),size(M,3),size(bootT,4));
tmp(1,:,:,:) = bootT; bootT = tmp;
clear tmp
end
else
bootT = squeeze(H0_data(:,:,1,:));
if size(M,1) == 1
tmp = NaN(1,size(M,2),size(bootT,3));
tmp(1,:,:) = bootT; bootT = tmp;
clear tmp
elseif size(M,2) == 1 %% for Weight bias testing
tmp = NaN(size(M,1),1,size(bootT,2));
tmp(:,1,:) = bootT; bootT = tmp;
clear tmp
end
end
[mask,M] = limo_max_correction(abs(M),abs(bootT),p,plotFlag); % threshold max absolute T values
if strncmp(FileName,'Rep_ANOVA_Interaction',21)
mytitle = sprintf('Interaction correction by T max');
elseif strncmp(FileName,'Rep_ANOVA_Gp_effect',19)
mytitle = sprintf('Gp effect correction by T max');
elseif strncmp(FileName,'Rep_ANOVA',9)
mytitle = sprintf('Main Effect correction by T max');
end
catch ME
errordlg(sprintf('error log: %s \n',ME.message),'max correction failure')
return
end
else
errordlg(['H0' filesep MCC_data ' not found'],'max correction failure')
return
end
% Correction using TFCE
% -------------------------------------
elseif MCC == 3 % Stat tfce
tfce_data = sprintf('tfce%stfce_%s',filesep, FileName);
H0_tfce_data = sprintf('H0%stfce_H0_%s', filesep, FileName);
if exist(tfce_data,'file') && exist(H0_tfce_data,'file')
try
tfce_data = load(tfce_data);
tfce_data = tfce_data.(cell2mat(fieldnames(tfce_data)));
H0_tfce_data = load(H0_tfce_data);
H0_tfce_data = H0_tfce_data.(cell2mat(fieldnames(H0_tfce_data)));
[mask,M] = limo_max_correction(tfce_data, H0_tfce_data,p,plotFlag);
if strncmp(FileName,'Rep_ANOVA_Interaction',21)
mytitle = sprintf('Interaction correction using TFCE');
elseif strncmp(FileName,'Rep_ANOVA_Gp_effect',19)
mytitle = sprintf('Gp effect correction using TFCE');
elseif strncmp(FileName,'Rep_ANOVA',9)
mytitle = sprintf('Main Effect correction using TFCE');
end
catch ME
limo_errordlg(sprintf('error log: %s \n',ME.message),'tfce correction failure')
return
end
else
errordlg('no tfce file or bootstrap file was found to compute the max distribution','missing data')
return
end
end
end