Experiments of 5-fold Cross-Validation on five kingdoms using Tree-augmented Naive Bayes

*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 2645 1
58
9
1
0 1
0 6
125
2846 .929
mit 2 1169 15 4
0 0
2
0 1
5
1198 .976
cyt 70
16 1686 30
3
2
4
2 15 17
1845 .914
ext 12
0 19
3668
2
0 5
8
136 93
3943
.930
gol 4
0 20
2
104 0 8
1
27
1
167 .623
pex 2
19
36
0 0 21
0 0 25
0
103 .204
end 5
1
8
1
4
0 425 0 11
2
457 .930
lys 0 0
5
14
0 0 0 126 24 1
170 .741
mem 20 10 101 52
2 0
52
3
4545 35
4820 .943
colSUM 2760 1216 1948 3780
116 23 497
140 4790 279
15549 O.R. = .925
precision .958 .961 .866 .970 .897 .913 .855 .900 .949 S.C. = .982
  O.P. = .942
specificity
.991
.997
.981
.990
.999
.999
.995
.999
.977


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 4
0
0
0 0 1 0 1
3
168 .946
mit 0
285 8
1 0 10 1 0 0 0 2
307 .928
cyt 1 3
432 1
0 8
0 0 0
0 2
447 .966
ext 0 0 4
110 0 0 0 0 1
0
12
127 .866
gol 0 0 0
0 16
0 0 0 0 18
1
35 .457
chl 1 17 25 2
0
1837 0
2 0
3
12
1899 .967
pex 0 2
12
1
0 3
12 0 0 0 0
29 .414
end 0 0 3
1
0 3
0 51 2
3 1
64 .797
vac 0 1
5
28
0
0 0 1
40
6
1
82 .488
mem 1
2
3
3
0
17
0 0
2
101
6
135 .748
colSUM 162 310
496 147 16
1877 13 55
45
132 40
3293 O.R. = .924
precision .981 .919 .871 .748 1.00
.979 .923 .927 .889 .765 S.C. = .988
  O.P. = .935
specificity
.999
.992
.978
.988
1.00
.971
1.00
.999
.998
.990


O.S. = .978

Table: Confusion matrix for Fungi Group
 
Prd ⇒
Obs ⇓
nuc  mit  cyt  ext  gol  pex  end  mem  vac  N.P.  rowSUM   recall 
nuc 532 3
15
2 0
0
0
3
0 66
621 .857
mit 14
305 35
2
0 0
0
7
0 43
406 .751
cyt 30 27 321 1
0
0
0
7
0
9
395 .813
ext 0 4
5
151 0
0
0
3
0
8
171 .883
gol 3
0 4
3
1
0
0
37
0
4
52 .019
pex 1
7
39
1
0 3
0 12
0 1
64 .047
end 4
1
11
1 0
0 11 32
0 4
64 .172
mem 6
2
11
4 0
0
0
269 0
10
302 .891
vac 0
0
5
11
0
0 0
3
0
0
19 .000
colSUM 590 349 446
176 1
3
11
373
0
145
2094  O.R. = .761
precision .902 .874 .720 .858 1.00
1.00
1.00
.721 -
S.C = .931
  O.P. = .817
specificity
.961
.974
.926
.987
1.00
.100
1.00
.942
1.00


O.S. = .954

Table: Confusion matrix for Gram-positive Bacteria group

Prd ⇒
Obs ⇓
cyt  wal  ext  mem  N.P.  rowSUM  recall 
cyt 897 0
17
7
9
930 .965
wal
1
0
15
1
2
19
.000
ext 33
0
194
7
18
252 .770
mem 14
0
16
287 23
340 .844
colSUM 945
0
242
302 52
1541  O.R. = .894
precision .949 -
.802 .950 S.C. = .966
  O.P. = .925
specificity
.921
1.00
.963
.988


O.S. = .942

Table: Confusion matrix for Gram-negative Bacteria group
 
Prd ⇒
Obs ⇓
cyt  ext  per  inn  wal  out  N.P.  rowSUM   recall 
cyt 1779 16
28 14 0 4 20
1861 .956
ext 3 215 8
0 0
3
24
253 .850
per 7
13
336 5
0 6 18
385
.873
inn 4
4
3
410 0 0 11
432 .949
wal
0 8
0 0 35 2
1
46 .761
out 2
1
2 2 0 182 8
197 .924
colSUM 1795 257
377 431 35 197 82
3174  O.R. = .932
precision .991 .837 .891 .951 1.00
.924 S.C. = .974
  O.P. = .956
specificity
.988
.986
.985
.992
1.00
.995


O.S. = .989

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 181 1
7
0 0
0 0 0 0 0 1
190 .953
ext 1 316 4
0 2
0 0
0 0 0 11
334 .946
nuc 0 2
332 0 5
0 0 0 0 0 13
352 .943
chl 2
0
5
81
2
0 0 0 0 0 4
94 .862
cyt 3
20
29
1 83 0 0 0 0 0 0
136 .610
end 2
8
0
1
0
0
0 3
0
0 0
14 0.00
lys 0 7
0
0 0 0 0
2
0 0 0
7 0.00
gol 0 4
0 0 1
0 0 17 0 0 0
22 .773
pex 1
0
2 0 5
0 0 0 0
0 0
8 .000
vac 1 3
0
0 0
0 0 0 0 0
0
4 0.00
colSUM 191 361 379 83
98
0
0
20
0
0
29
1161 O.R. = .870
precision .948 .875 .876 .876 .847 -
-
.850 -
-
S.C. = .975
  O.P. = .892
specificity
.990
.946
.942
.998
.985
1.00
1.00
.997
1.00
1.00


O.S. = .960

Table: Confusion matrix for PSORT-B data

Prd ⇒
Obs ⇓
cyt  inn  per  out  ext  N.P.  rowSUM  recall 
cyt 208 14
11
0
9
10
252 .825
inn 4
281 4
1 0
18
308 .912
per 5
19
219 5
11
14
264 .830
out 3
0
1 358 7
9
378 .947
ext 1
0
5
1
211 23
241 .876
colSUM 239 293 241 368 228
74
1443 O.R. = .885
precision .941 .921 .912 .981 .887 S.C. = .949
  O.P. = .933
specificity
.989
.979
.982
.993
.978


O.S. = .985