-
Notifications
You must be signed in to change notification settings - Fork 1
/
runCV_NIDM.m
150 lines (134 loc) · 7.09 KB
/
runCV_NIDM.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
% % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % %
% % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % %
datadir = 'data';
ResPerfDir = 'results'; if ~exist(ResPerfDir,'dir');mkdir(ResPerfDir); end
load([datadir,filesep,'demoDataset.mat'])
AdjDfD = speye( size(AdjGfD,2) );
CoEXPname = '';
SubNetworkNames = fieldnames(MatrixSet_gg )';
fdatestr = datestr(now,'yyyy.mmm.dd-HH.MM.SS') ;
disp(['Net:',netname, 'Date:',fdatestr]);
disp( SubNetworkNames' )
% % % % % % % % % % % % % % % % % % % % % % % % % % % %
rng(1111)
nCVTimes = 5; nFold = 5; CVtype = [num2str(nFold),'FCV']; MinSizeDisGeSet = nFold;
%
n_disgenes_eachdisease = sum(Matrix_gene_dis,1)';
dis_IDset = find(n_disgenes_eachdisease>=MinSizeDisGeSet);
% dis_IDset = dis_IDset(1:2) %%%for test only %%%%%%%%%%%
n_neighbors_artificial_linkage_interval = 99;
%
n_disease_in_Table = length( dis_IDset );
nCV_list = zeros( n_disease_in_Table, 1 );
matAUROC_nCVTimes = cell(nCVTimes,1);
matAUPRC_nCVTimes = cell(nCVTimes,1);
matAURecall_nCVTimes = cell(nCVTimes,1);
matAUPrec_nCVTimes = cell(nCVTimes,1);
matRec_nCVTimes = cell(nCVTimes,1);
matPrec_nCVTimes = cell(nCVTimes,1);
methodset_nCVTimes = cell(nCVTimes,1);
date_start = datestr(now,'yyyy.mmm.dd-HH.MM.SS');
for i_cv = 1:nCVTimes
disp(['i_cv-',num2str(i_cv) ])
MatrixSet_gg_copy= MatrixSet_gg;
%
matAUROC = [];
matAUPRC = [];
matAURecall = [];
matAUPrec = [];
matRec = [];
matPrec = [];
methodset = [] ;
idx_res = 0;
for ii_dis = 1:n_disease_in_Table
tic
Matrix_gene_dis_copy = Matrix_gene_dis ;
ID_dis = dis_IDset(ii_dis);
disp(['i_cv-',num2str(i_cv),'; ii_dis-',num2str(ii_dis),'; ID_dis-',num2str(ID_dis)])
ac_gene_dis00 = Matrix_gene_dis_copy(:,ID_dis );
idx_pos = find( ac_gene_dis00 ); n_pos = length( idx_pos);
idx_neg = find( ~ac_gene_dis00 ); n_neg = length( idx_neg);
n_fold_real = min(nFold, n_pos) ;
ind_fold_pos = crossvalind('Kfold', n_pos, n_fold_real ) ;
%ind_fold_neg = crossvalind('Kfold', n_neg, n_fold_real ) ;
for i_fold = 1:n_fold_real
% idx_pos_train = idx_pos(ind_fold_pos~=i_fold);
idx_pos_test = idx_pos(ind_fold_pos==i_fold); n_pos_test =length(idx_pos_test);
%
idx_neg_test_X100_RC = idx_neg( randperm(n_neg, min(n_pos_test*n_neighbors_artificial_linkage_interval,n_neg) ) ); n_neg_test_x100 = length( idx_neg_test_X100_RC ) ; % 每个测试基因选择100个对应控制基因，总共n_pos_test*100
idx_test_pos_neg_X100_RC = [idx_neg_test_X100_RC; idx_pos_test ] ;
%
idx_neg_test_WG = idx_neg ; n_neg_test_all = length( idx_neg_test_WG ) ;
idx_test_pos_neg_WG = [idx_neg_test_WG; idx_pos_test ] ;
%
AdjGfD = Matrix_gene_dis_copy;
AdjGfD(idx_pos_test,ID_dis ) = 0 ;
P0 = AdjGfD(:,ID_dis );
%
%
TableScores = table ;
% M
[TableScores_NIDM ] = A_NIDM(MatrixSet_gg_copy, P0 ) ;
TableScores_NIDM.Properties.VariableNames = strcat( TableScores_NIDM.Properties.VariableNames, [CoEXPname] );
TableScores = [TableScores,TableScores_NIDM];
% A_NIDMgss_v2 is preferred
% no gSemSim
[TableScores_NIDM ] = A_NIDMgss_v2(MatrixSet_gg_copy,[],[], P0, [], [] );
TableScores_NIDM.Properties.VariableNames = strcat( TableScores_NIDM.Properties.VariableNames, [CoEXPname] );
TableScores = [TableScores,TableScores_NIDM];
% +gss
[TableScores_NIDM ] = A_NIDMgss_v2(MatrixSet_gg_copy,AdjGfD,AdjDfD, P0, [], [] );
TableScores_NIDM.Properties.VariableNames = strcat( TableScores_NIDM.Properties.VariableNames, [CoEXPname] );
TableScores = [TableScores,TableScores_NIDM];
% % % % % % % % % % % % % % % % % % % % %
methodset = TableScores.Properties.VariableNames ;
test_real = ac_gene_dis00(idx_test_pos_neg_WG);
[AUROCset, AUPRCset , Rec5set , Prec5set, AURecallset, AUPrecset ] = getPerf(test_real,TableScores(idx_test_pos_neg_WG,:) ) ;
%
test_real_RC = ac_gene_dis00(idx_test_pos_neg_X100_RC);
[AUROCset_RC, AUPRCset_RC , Recset_RC , Precset_RC, AURecallset_RC, AUPrecset_RC ] = getPerf(test_real_RC,TableScores(idx_test_pos_neg_X100_RC,:) ) ;
idx_res = idx_res +1 ;
matAUROC(idx_res,:) = [AUROCset_RC, AUROCset ];
matAUPRC(idx_res,:) = [AUPRCset_RC, AUPRCset ];
matAURecall(idx_res,:) = [AURecallset_RC, AURecallset ];
matAUPrec(idx_res,:) = [AUPrecset_RC, AUPrecset ];
matRec(idx_res,:) = [Recset_RC, Rec5set ];
matPrec(idx_res,:) = [Precset_RC, Prec5set ];
% % % % %
end
end
%
matAUROC_nCVTimes{i_cv} = mean(matAUROC,1);
matAUPRC_nCVTimes{i_cv} = mean(matAUPRC,1) ;
matAURecall_nCVTimes{i_cv}= mean(matAURecall,1);
matAUPrec_nCVTimes{i_cv} = mean(matAUPrec,1);
matRec_nCVTimes{i_cv} = mean(matRec,1);
matPrec_nCVTimes{i_cv} = mean(matPrec,1);
methodset_nCVTimes{i_cv} = methodset;
toc
disp(' ')
mean(matAUPRC,1)
end
% % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % % %
%
matAUROC_nCVTimes = cat(1,matAUROC_nCVTimes{:});
matAUPRC_nCVTimes = cat(1,matAUPRC_nCVTimes{:});
matAURecall_nCVTimes = cat(1,matAURecall_nCVTimes{:});
matAUPrec_nCVTimes = cat(1,matAUPrec_nCVTimes{:});
matRec_nCVTimes = cat(1,matRec_nCVTimes{:});
matPrec_nCVTimes = cat(1,matPrec_nCVTimes{:});
%
vars = [strcat( methodset_nCVTimes{1}, '_RC'), methodset_nCVTimes{1} ] ;
matRESmean = [ mean(matAUROC_nCVTimes,1);...
mean(matAUPRC_nCVTimes,1);...
mean(matAURecall_nCVTimes,1);...
mean(matAUPrec_nCVTimes,1);...
mean(matRec_nCVTimes,1); ...
mean(matPrec_nCVTimes,1) ] ;
tbRESmean = array2table(matRESmean, 'VariableNames', vars, 'RowNames',{'AUROC','AUPRC','AURecall','AUPrec','Rec','Prec'})
%
% save
date_cmplt = datestr(now,'yyyy.mmm.dd-HH.MM.SS');
parastr = sprintf('CVtype=%s_CVtime=%d_MSDGS%d', CVtype , nCVTimes, MinSizeDisGeSet );
outfile = [ResPerfDir,filesep,'ResPerf_NIDM_',PPItype,'_',parastr,'_',date_cmplt,'.mat']
save([outfile], 'tbRESmean', 'date_start', 'date_cmplt' , '-v7.3' ) ;