Table 1 GenDev Technique A: Results for 2400 computational experiments using the General Development Mode (GenDev) with each biomass metabolite as an independent part of the objective function, that is, no constraint is applied to the overall biomass reaction
From: Evaluation of reaction gap-filling accuracy by randomization
(1) # rxns removed | 1 | 2 | 5 | 10 | 20 | 50 | ||||||
(2) Rxns suggested are identical to rxns removed | 48 | na | 22 | na | 3 | na | 0 | na | 0 | na | 0 | na |
52 | 50 | 22 | 23 | 3 | 7 | 0 | 0 | 0 | 0 | 0 | 0 | |
(3) No rxns suggested | 36 | na | 15 | na | 0 | na | 0 | na | 0 | na | 0 | na |
36 | 25 | 17 | 13 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
(4) Rxns suggested are a strict subset of rxns removed | 0 | na | 40 | na | 43 | na | 27 | na | 8 | na | 0 | na |
0 | 0 | 42 | 35 | 40 | 7 | 21 | 36 | 9 | 23 | 0 | 4 | |
(5) # rxns suggested is equal to # rxns re-moved | 11 | na | 9 | na | 11 | na | 4 | na | 0 | na | 0 | na |
9 | 18 | 8 | 20 | 13 | 51 | 7 | 4 | 2 | 0 | 0 | 0 | |
(6) # rxns suggested is less than # rxns re-moved | 0 | na | 8 | na | 40 | na | 64 | na | 91 | na | 98 | na |
0 | 0 | 9 | 5 | 37 | 6 | 63 | 60 | 89 | 77 | 99 | 96 | |
(7) # rxns suggested is greater than # rxns removed | 5 | na | 6 | na | 3 | na | 5 | na | 1 | na | 0 | na |
3 | 7 | 2 | 4 | 6 | 33 | 9 | 0 | 0 | 0 | 0 | 0 | |
(8) # cases where no solution was found | 0 | na | 0 | na | 0 | na | 0 | na | 0 | na | 2 | na |
0 | 0 | 0 | 0 | 0 | 3 | 0 | 0 | 0 | 0 | 1 | 0 | |
Precision (%) | 77 | na | 81 | na | 79 | na | 80 | na | 79 | na | 77 | na |
83 | 69 | 83 | 78 | 75 | 84 | 79 | 85 | 79 | 89 | 76 | 89 | |
Average precision (%) | 79 | na | ||||||||||
79 | 82 | |||||||||||
Recall (%) | 52 | na | 51 | na | 54 | na | 52 | na | 52 | na | 52 | na |
54 | 54 | 48 | 51 | 50 | 59 | 57 | 57 | 57 | 59 | 57 | 59 | |
Average recall (%) | 52 | na | ||||||||||
54 | 56 | |||||||||||
193 (32%) | na | |||||||||||
223 (37%) | 144 (24%) | |||||||||||