/
s_fe_weights_hcp.m
315 lines (272 loc) · 12.9 KB
/
s_fe_weights_hcp.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
function s_fe_weights_hcp(bval,tractographyType)
%
% This function:
% - Load a series of results precomputed by s_fe_save_culled_fg(bval)
% - Extract the weights and fiber lenths.
% - Make a plot of the distribution of weights and fiber lengths.
%
% s_fe_weights(bval,tractographyType)
%
% Copyright Franco Pestilli (2014) Stanford University
% Get the base directory for the data
datapath = '/marcovaldo/frk/2t2/HCP/';
subjects = {'105115','110411','111312','113619','115320','117122','118730'};
if notDefined('tractographyType'), tractographyType = 'lmax10'; end
if notDefined('bval'), bval = 2000; end
if notDefined('saveDir'), saveDir = fullfile('/marcovaldo/frk/Dropbox','pestilli_etal_revision',mfilename);end
% These are default parameters for plotting
nBins = 11; % first bin is zero weight second is
bins_w = linspace(-10,-1,nBins);
proportionDeleted = nan(size(subjects,2),2);
w = cell(1,length(subjects));
len = w;
for isbj = 1:length(subjects)
% Directory where to load the results
if isbj <= 4
datapath = '/marcovaldo/frk/2t1/HCP/';
else
datapath = '/marcovaldo/frk/2t2/HCP/';
end
resultsPath = fullfile(datapath,subjects{isbj},'results');
savedir = fullfile(saveDir,subjects{isbj});
if ~isempty(bval)
resFiles = dir(fullfile(resultsPath,sprintf('*%s*%s*cerebellum*.mat',num2str(bval),tractographyType)));
else
resFiles = dir(fullfile(resultsPath,sprintf('*%s*right*.mat',tractographyType)));
end
% We build one modelper fiber group, whole brain fiber group
for iRes = 1:length(resFiles)
% The final connectome and dat astructure will be saved with this name:
[~,resFileName] = fileparts(resFiles(iRes).name);
% Buil a full-file of the fibers and the RESULTS structure to load
resFileName2Load = fullfile(resultsPath,[resFileName,'.mat']);
% Initialize the Connectome
fprintf('[%s] Loading results: \n%s\n',mfilename,resFileName2Load)
load(resFileName2Load);
% Reorganize the weights and fiber-lengths in structures divided by
% subjects
w{isbj} = results.weights(results.weights > 0);
olen{isbj} = results.optimized.length;
clen{isbj} = results.candidate.length;
% Distribution of length
[f, ol(isbj,:), cl(isbj,:), lbins] = plotLengthHist(tractographyType,olen{isbj},clen{isbj});
feSavefig(f,'verbose','yes','figName',[resFileName, 'cereb_optimized_fiberlength_hist'],'figDir',savedir,'figType','eps');
close(f)
% Proportion of deleted fibers at each fiber length
rl = ol(isbj,:)./cl(isbj,:);
drl = nanmean(rl,1);
f = plotLengthRatioHist(tractographyType,drl,lbins);
feSavefig(f,'verbose','yes','figName',[resFileName, 'cereb_length_ratio_hist'],'figDir',savedir,'figType','eps');
close(f)
% Distribution of weights
[f, ow(isbj,:), wbins] = plotWeightsHist(tractographyType,w{isbj});
feSavefig(f,'verbose','yes','figName',[resFileName, 'cereb_optimized_weight_hist'],'figDir',savedir,'figType','eps');
close(f)
% 2d histogram
%[f, h] = plotLengthWeightScatter(tractographyType,w{isbj},len{isbj});
%feSavefig(f,'verbose','yes','figName',[resFileName, '_weight_vs_fiberlength_2d'],'figDir',savedir,'figType','eps');
%set(gca, 'visible', 'off')
%feSavefig(f,'verbose','yes','figName',[resFileName, '_weight_vs_fiberlength_2d'],'figDir',savedir,'figType','jpg');
%close(f)
% Just reorganize the vectors
%len{isbj} = len{isbj}';
%w{isbj} = w{isbj}';
% Save the proportion of zero-weight and non-zero-weight fibers
proportionDeleted(isbj,:) = [results.candidate.n - results.optimized.n results.optimized.n]./results.candidate.n;
[f, h] = plotFiberPie(resFileName,proportionDeleted(isbj,:));
feSavefig(f,'verbose','yes','figName',[resFileName, 'cereb_pie'],'figDir',savedir,'figType','eps');
close(f)
end
end
mow = mean(ow,1);
owerr = [mow;mow]+[-std(ow)./sqrt(size(ow,1));std(ow)./sqrt(size(ow,1))];
f = plotMeanWeightsHist(tractographyType,mow,owerr,wbins);
feSavefig(f,'verbose','yes','figName',[resFileName, 'cereb_pie_weights'],'figDir',saveDir,'figType','eps');
close(f)
mol = 2*mean(ol,1);
olerr = [mol;mol]+[-std(ol)./sqrt(size(ow,1));std(ol)./sqrt(size(ow,1))];
mcl = 2*mean(cl,1);
clerr = [mcl;mcl]+[-std(cl)./sqrt(size(ow,1));std(cl)./sqrt(size(ow,1))];
f = plotMeanLengthHist(tractographyType,mol,olerr,mcl,clerr,lbins);
feSavefig(f,'verbose','yes','figName',[resFileName, 'cereb_hist_length'],'figDir',saveDir,'figType','eps');
close(f)
[f, h] = plotProportionDeleted(tractographyType,proportionDeleted);
feSavefig(f,'verbose','yes','figName',['mean_',tractographyType, 'cereb_deleted_fibers_across_subjects'],'figDir',saveDir,'figType','eps');
close(f)
ratioLength = ol./cl;
drl = nanmean(ratioLength,1);
drlerr = 10.^(log10([drl;drl])+[-nanstd(ratioLength)./sqrt(size(ratioLength,1));nanstd(ratioLength)./sqrt(size(ratioLength,1))]);
f = plotMeanLengthRatioHist(tractographyType,drl,drlerr,lbins);
feSavefig(f,'verbose','yes','figName',[tractographyType, 'cereb_length_ratio_hist_across_subjects'],'figDir',fullfile(saveDir,'averages'),'figType','eps');
close(f)
% [f, h] = plotLengthWeightScatter(['_ACROSS_FIVE_SUBJECTS_',tractographyType],cell2mat(w)',cell2mat(len)');
% feSavefig(f,'verbose','yes','figName',[tractographyType, '_weight_vs_fiberlength_2d_across_subjects'],'figDir','~/2t1/predator/average_figures','figType','eps');
% set(gca, 'visible', 'off')
% feSavefig(f,'verbose','yes','figName',[tractographyType, '_weight_vs_fiberlength_2d_across_subjects'],'figDir','~/2t1/predator/average_figures','figType','jpg');
% close(f)
end % End main function
function f = plotMeanLengthRatioHist(tractographyType,drl,drlerr,bins)
fontSiz = 15;
% Make a 2D histogram (scatter plot) of the percent deleted and kept fibers
f = figure('name',sprintf('mean_length_ratio_hist_%s',tractographyType),'color','w');
sh = plot([bins;bins],drlerr,'-','color','r','linewidth',2);
hold on
sh = plot(bins,(drl),'-','color','k','markeredgecolor','w','markerfacecolor','k','markersize',12);
ylabel('Proportion deleted fascicles','fontsize',fontSiz)
xlabel('Fascicle length','fontsize',fontSiz)
set(gca, 'ylim', [0.125/8 .6], ...
'ytick',[0.125/8 0.125/4 0.125/2 0.125 0.25 0.5], ...
'xlim',[bins(1) bins(end)],...
'xtick',[bins(1) 8 32 128 bins(end)],...
'yscale','log', ...
'xscale','log', ...
'tickdir','out','box','off', ...
'fontsize',fontSiz,'visible','on')
end
function f = plotLengthRatioHist(tractographyType,drl,bins)
fontSiz = 15;
% Make a 2D histogram (scatter plot) of the percent deleted and kept fibers
f = figure('name',sprintf('length_ratio_hist_%s',tractographyType),'color','w');
sh = plot(bins,(drl),'-','color','k','markeredgecolor','w','markerfacecolor','k','markersize',12);
ylabel('Proportion deleted fascicles','fontsize',fontSiz)
xlabel('Fascicle length','fontsize',fontSiz)
set(gca,'xlim',[0 200],...
'xtick',[0 100 200],...
'ylim', [0.125/8 .5], ...
'ytick',[0.125/8 0.125/4 0.125/2 0.125 0.25 0.5], ...
'yscale','log', ...
'tickdir','out','box','off', ...
'fontsize',fontSiz,'visible','on')
end
function f = plotMeanLengthHist(tractographyType,ol,olerr,cl,clerr,bins)
fontSiz = 15;
% Make a 2D histogram (scatter plot) of the percent deleted and kept fibers
f = figure('name',sprintf('mean_length_hist_%s',tractographyType),'color','w');
patch([bins,fliplr(bins)],[cl,zeros(size(cl))],[.6 .6 .6]);
hold on
patch([bins,fliplr(bins)],[ol,zeros(size(ol))],[.4 .4 .4]);
sh = plot([bins;bins],olerr,'-','color','r','linewidth',2);
sh = plot([bins;bins],clerr,'-','color','r','linewidth',2);
ylabel('Number of fascicles','fontsize',fontSiz)
xlabel('Fascicle length (mm)','fontsize',fontSiz)
set(gca,'xscale','log', ...
'xlim',[2 256],...
'xtick',[2 4 8 16 32 64 128 256],...
'ylim', [0 12000], ...
'ytick',[0 6000 12000], ...
'tickdir','out','box','off', ...
'fontsize',fontSiz,'visible','on')
end
function f = plotMeanWeightsHist(tractographyType,ow,owerr,bins)
fontSiz = 15;
% Make a 2D histogram (scatter plot) of the percent deleted and kept fibers
f = figure('name',sprintf('mean_weights_hist_%s',tractographyType),'color','w');
sh = plot([bins;bins],owerr,'-','color','r','linewidth',3);
hold on
sh = plot(bins,(ow),'-','color','k','markeredgecolor','w','markerfacecolor','k','markersize',12);
ylabel('Number of fascicles','fontsize',fontSiz)
xlabel('log_{10}(Fascicle weight)','fontsize',fontSiz)
set(gca,'xlim',[-6 0],...
'xtick',[-6 -3 0],...
'ylim', [0 8000], ...
'ytick',[0 4000 8000], ...
'tickdir','out','box','off', ...
'fontsize',fontSiz,'visible','on')
end
function [f, oy,bins] = plotWeightsHist(tractographyType,ow)
fontSiz = 15;
% Make a 2D histogram (scatter plot) of the percent deleted and kept fibers
f = figure('name',sprintf('weights_hist_%s',tractographyType),'color','w');
bins = linspace(-6,0,22);
[oy,x] = hist(log10(ow),bins);
sh = plot(bins,oy,'-','color','k','markeredgecolor','w','markerfacecolor','k','markersize',12);
ylabel('Number of fascicles','fontsize',fontSiz)
xlabel('log_{10}(Fiber weight)','fontsize',fontSiz)
set(gca,'xlim',[-6 0],...
'xtick',[-6 -3 0],...
'ylim', [0 8000], ...
'ytick',[0 4000 8000], ...
'tickdir','out','box','off', ...
'fontsize',fontSiz,'visible','on')
end
function [f, oy, cy, bins] = plotLengthHist(tractographyType,olen,clen)
fontSiz = 15;
% Make a 2D histogram (scatter plot) of the percent deleted and kept fibers
f = figure('name',sprintf('fiberlength_hist_%s',tractographyType),'color','w');
%bins = [2,4,8,16,32,64,128,256,512];
bins = [2:2:512];
[oy,x] = hist(olen,bins);
[cy,x] = hist(clen,bins);
patch([x,fliplr(x)],[cy,zeros(size(cy))],[.6 .6 .6]);
hold on
patch([x,fliplr(x)],[oy,zeros(size(oy))],[.4 .4 .4]);
ylabel('Number of fascicles','fontsize',fontSiz)
xlabel('Fascicle length (mm)','fontsize',fontSiz)
set(gca,'xscale','log', ...
'xlim',[2 256],...
'xtick',[2 4 8 16 32 64 128 256],...
'ylim', [0 10000], ...
'ytick',[0 5000 10000], ...
'tickdir','out','box','off', ...
'fontsize',fontSiz,'visible','on')
end
function [f, ymap] = plotLengthWeightScatter(tractographyType,w,len)
fontSiz = 15;
% Make a 2D histogram (scatter plot) of the percent deleted and kept fibers
f = figure('name',sprintf('weight_vs_fiberlength_%s',tractographyType),'color','w');
nBins = 46;
ybins = linspace(0,nBins,nBins);
ybins = logspace(log10(5),log10(400),nBins);
xbins = linspace(-9,-1,nBins);
[ymap,x] = hist3([(len), log10(w)],{ ybins, xbins});
ymap = ymap./numel(len);
sh = imagesc(flipud(ymap));
cm = colormap(flipud(hot));
view(0,90); axis('square');
xlabel('log_1_0(Fascicle weight)','fontsize',fontSiz)
ylabel('Fascicle length (mm)','fontsize',fontSiz)
set(gca,'xlim',[nBins/2 nBins],...
'ylim', [5 nBins], ...
'ytick',[5 nBins/2 nBins], ...
'yticklabel',[round(ybins(nBins)) round(ybins(round(mean([nBins, nBins/2])))) round(ybins(nBins/2)) ], ...
'xtick',[nBins/2 mean([nBins, nBins/2]) nBins], ...
'xticklabel',[round(xbins(nBins/2)) round(xbins(round(mean([nBins, nBins/2])))) round(xbins(nBins)) ], ...
'tickdir','out','box','off', ...
'fontsize',fontSiz,'visible','on')
end
function [f, h] = plotProportionDeleted(tractographyType,proportionDeleted)
% Make aplot of the percent deleted and kept fibers
fontSiz = 15;
f = figure('name',sprintf('mean_proportion_deleted_fibers_%s',tractographyType),'color','w');
m = mean(proportionDeleted,1);
sd = [m - std(proportionDeleted)./sqrt(size(proportionDeleted,1)); m + std(proportionDeleted./sqrt(size(proportionDeleted,1)))];
h(1) = bar(m,'facecolor','k'); hold on
plot([1 2; 1 2],sd,'r-','linewidth',4)
ylabel('Proportion','fontsize',fontSiz)
set(gca,'tickdir','out','color','w','box','off', ...
'ylim',[0 1], ...
'ytick',[0 .25 .5 .75 1], ...
'xlim',[0 3], 'fontsize',fontSiz,...
'xticklabel',{'Deleted','Kept'})
end
function [f, h] = plotFiberPie(resFileName,proportionDeleted)
% Make a pie graphshwing the percent of values below and above zero
fontSiz = 15;
f = figure('name',sprintf('pie_of_deleted_fibers_%s',resFileName),'color','w');
h = pie( proportionDeleted,[1 0]);
colormap gray
patchObjs = findobj(h,'Type','patch');
textObjs = findobj(h,'Type','text');
oldStr = get(textObjs,{'String'});
val = get(textObjs,{'Extent'});
oldExt = cat(1,val{:});
newStr = {sprintf('%s deleted',oldStr{1});sprintf('%s kept',oldStr{2})};
set(textObjs,{'String'},newStr)
end
% load run01_fliprot_aligned_trilin_csd_lmax10_run01_fliprot_aligned_trilin_brainmask_run01_fliprot_aligned_trilin_wm_prob-800000-fiberStatsResults.mat
% w = results.weights(results.weights>0);
% [f,x_vals] = ecdf(log10(w));
% F = plot(x_vals,f);
% hold on
% G = plot(x_vals,normcdf(x_vals,mean(log10(w)),std(log10(w))),'r-');
% [H, pValue, KSstatistic, criticalValue] = kstest(f)