Experiments of 5-fold Cross-Validation on five kingdoms using Support Vector Machines

*For ontology and other abbreviations, see here

Table 1: Confusion matrix for Animal group

Prd ⇒
Obs ⇓
nuc  mit  cyt  ext  gol  pex  end  lys  mem  N.P.  rowSUM  recall 
nuc 2564 12 101
10 9
2
6
1 16
125
2846 .901
mit 1
1175 8
1
0
0
0
0 8
5
1198 .981
cyt 4
2
1770 12
2
2
1
1
34 17
1845 .959
ext 3
0 5
3722
0
0 1
7
112
93
3943
.944
gol 0
0 6
0
158
0 2
0
0
1
167 .946
pex 0 1
0
0 0 100
0 0 2
0
103 .971
end 0 2
2
0
1 0 450
0
0
2
457 .985
lys 0 0
0
1
1
0 0 160 7
1
170 .941
mem 3
1 14 4
1
0
0
0
4762 35
4820 .988
colSUM 2575
1193
1906
3750
172
104
460
168
4941
279
15549 O.R. = .956
precision .996
.985
.929
.993
.917
.962
.978
.952
.964
S.C. = .982
  O.P. = .973
specificity
.999
.990
.990
.998
.999
.999
.999
.999
.983


O.S. = .997

Table 2: Confusion matrix for Plant group

Prd ⇒
Obs ⇓
nuc  mit  cyt  ext  gol  chl  pex  end  vac  mem  N.P.  rowSUM   recall 
nuc 159
0
0
1
1
2
0
2
0
0
3
168 .946
mit 0
285
2
0
0
17
1
0
0
0
2
307 .928
cyt 1
1
422
6
0
14
0
0
1
0
2
447 .944
ext 0
1
1
106
0
2
0
0
4
1
12
127 .835
gol 0
0
0
0
34
0
0
0
0
0
1
35 .971
chl 1
14
8
5
0
1854
1
2
0
2
12
1899 .976
pex 0
0
0
0
0
1
28
0
0
0
0
29 .966
end 1
0
0
0
0
2
0
54
3
3
1
64 .844
vac 0
0
1
3
0
1
0
2
72
2
1
82 .878
mem 0
0
2
4
0
7
0
3 7
106
6
135 .785
colSUM 162
301
436
125
35
1900
30
63
87
114
40
3293 O.R. = .947
precision .981
.947
.968
.848
.971
.976
.933
.857
.828
.930
S.C. = .988
  O.P. = .959
specificity
.999
.995
.995
.994
.999
.967
.999
.997
.995
.997


O.S. = .997

Table: Confusion matrix for Fungi Group
 
Prd ⇒
Obs ⇓
nuc  mit  cyt  ext  gol  pex  end  mem  vac  N.P.  rowSUM   recall 
nuc 515
7
23
2
1
0
1
6
0
66
621 .887
mit 13
319
20
3
0
3
1
3
1
43
406 .786
cyt 37
22
312
4
1
4
2
2
2
9
395 .790
ext 0
3
4
145
1
0
1
5
4
8
171 .848
gol 3
0
0
0
39
0
2
4
0
4
52 .750
pex 2
5
1
0
0
55
0
0
0
1
64 .859
end 4
0
1
2
3
0
42
8
0
4
64 .656
mem 6
1
4
4
3
0
4
268
2
10
302 .887
vac 1
0
4
3
1
0
0
0
10
0
19 .526
colSUM 581
357
369
163
49
62
53
296
19
145
2094  O.R. = .814
precision .886
.894
.846
.890
.796
.887
.792
.905
.526
S.C = .931
  O.P. = .875
specificity
.955
.977
.966
.990
.995
.997
.995
.984
.996


O.S. = .995

Table: Confusion matrix for Gram-positive Bacteria group

Prd ⇒
Obs ⇓
cyt  wal  ext  mem  N.P.  rowSUM  recall 
cyt 903
0
6
12
9
930 .979
wal
1
15
1
0
2
19
.938
ext 5
1
214
14
18
252 .877
mem 13
0
23
281
23
340 .826
colSUM 922
16
244
307
52
1541  O.R. = .898
precision .971
.789
.849
.915
S.C. = .966
  O.P. = .929
specificity
.969
.999
.977
.978


O.S. = .988

Table: Confusion matrix for Gram-negative Bacteria group
 
Prd ⇒
Obs ⇓
cyt  ext  per  inn  wal  out  N.P.  rowSUM   recall 
cyt 1816
6
15
2
0
2
20
1861 .976
ext 15
199
9
0
2
4
24
253 .787
per 21
8
330
5
0
3
18
385
.857
inn 9
0
1
411
0
0
11
432 .951
wal
0
1
0
0
43
1
1
46 .935
out 2
3
3
0
0
181
8
197 .919
colSUM 1863
217
358
418
45
191
82
3174  O.R. = .939
precision .975
.917
.917
.983
.956
.948
S.C. = .974
  O.P. = .963
specificity
.964
.994
.990
.997
.999
.997


O.S. = ..997

Table: Confusion matrix for Nair & Rost (1161 proteins in Swiss-Prot)

Prd ⇒
Obs ⇓
mit  ext  nuc  chl  cyt  end  lys  gol  pex  vac  N.P.  rowSUM  recall 
mit 186
1
0
0
0
1
0
0
1
0
1
190 .979
ext 1
314
1
0
4
1
1
1
0
0
11
334 .940
nuc 1
5
322
0
11
0
0
0
0
0
13
352 985
chl 0
0
0
89
1
0
0
0
0
0
4
94 .947
cyt 3
2
9
1
120
1
0
0
0
0
0
136 .882
end 0
0
0
0
0
13
0
1
0
0
0
14 .929
lys 0
0
1
0
0
0
6
0
0
0
0
7 .857
gol 0
4
1
0
0
0
0
17
0
0
0
22 .773
pex 0
1
1
0
0
0
0
0
6
0
0
8 .750
vac 1
0
2
0
1
0
0
0
0
0
0
4 0
colSUM 192
327
337
90
137
16
7
19
7
0
29
1161 O.R. = .924
precision .969
.960
.955
.989
.876
.812
.857
.895
.857
-
S.C. = .975
  O.P. = .948
specificity
.993
.984
.981
.999
.983
.997
.999
.998
.999



O.S. = .989

Table: Confusion matrix for PSORT-B data

Prd ⇒
Obs ⇓
cyt  inn  per  out  ext  N.P.  rowSUM  recall 
cyt 221
3
8
1
9
10
252 .877
inn 4
279
3
1
3
18
308 .906
per 10
4
221
6
9
14
264 .837
out 2
1
1
361
4
9
378 .955
ext 1
1
10
7
199
23
241 .826
colSUM 238
288
243
376
224
74
1443 O.R. = .888
precision .929
.969
.909
.960
.888
S.C. = .949
  O.P. = .936
specificity
.986
.992
.981
.986
.979


O.S. = .989