Table 4 |
||||||||||
|
Performance comparison of MultiLoc2-HighRes with WoLF PSORT using Höglund independent dataset (Höglund IDS) |
||||||||||
|
Animals |
Fungi |
Plants |
||||||||
|
Predictor |
Loc |
No. |
SE |
SE3 |
No. |
SE |
SE3 |
No. |
SE |
SE3 |
|
|
||||||||||
|
MultiLoc2-HighRes |
ex |
78 |
78 |
91 |
7 |
77 |
86 |
1 |
0 |
100 |
|
pm |
34 |
55 |
78 |
29 |
10 |
31 |
6 |
33 |
50 |
|
|
pe |
3 |
33 |
100 |
5 |
20 |
100 |
2 |
50 |
100 |
|
|
er |
25 |
28 |
70 |
46 |
46 |
83 |
6 |
50 |
83 |
|
|
go |
14 |
7 |
57 |
8 |
25 |
63 |
6 |
33 |
50 |
|
|
ly |
4 |
25 |
75 |
- |
- |
- |
- |
- |
- |
|
|
va |
- |
- |
- |
11 |
0 |
0 |
9 |
11 |
33 |
|
|
AVG |
38 |
79 |
30 |
61 |
30 |
69 |
||||
|
ACC |
57 |
82 |
31 |
59 |
30 |
57 |
||||
|
|
||||||||||
|
WoLF PSORT |
ex |
78 |
93 |
97 |
7 |
36 |
79 |
1 |
0 |
0 |
|
pm |
34 |
41 |
59 |
29 |
59 |
79 |
6 |
83 |
83 |
|
|
pe |
3 |
0 |
0 |
5 |
0 |
0 |
2 |
0 |
0 |
|
|
er |
25 |
8 |
40 |
46 |
9 |
54 |
6 |
0 |
50 |
|
|
go |
14 |
0 |
7 |
8 |
0 |
0 |
6 |
17 |
17 |
|
|
ly |
4 |
0 |
25 |
- |
- |
- |
- |
- |
- |
|
|
va |
- |
- |
- |
11 |
0 |
0 |
9 |
0 |
33 |
|
|
AVG |
24 |
38 |
17 |
35 |
17 |
31 |
||||
|
ACC |
58 |
68 |
22 |
51 |
20 |
40 |
||||
|
|
||||||||||
|
The sensitivity (SE) and top three sensitivity (SE3) for each localization are shown. SE3 measures the fraction of correctly predicted proteins within the top three ranked localizations. The corresponding average sensitivity and overall accuracy are listed also, with the top-scoring highlighted in bold. Values based on very few proteins (less than six) are drawn in gray. All measures are given as percentages. |
||||||||||
|
Blum et al. BMC Bioinformatics 2009 10:274 doi:10.1186/1471-2105-10-274 |
||||||||||
