Network
dataset
w
1
(PFP-GF)
w
2
(B2G-GF)
w
3
(EFICAz-GF)
Θ
iIN800
0.60
0.10
0.30
0.63
iWV1314
0.43
0.47
0.10
0.40
iHD666
0.65
0.20
0.15
0.63
iMA871
0.40
0.50
0.10
0.40
iIB711
0.13
0.30
0.57
0.30
Values of parameters (w
i
, Θ) optimized by maximizing F
2
score on the known dataset N
K
for each network. These weights and thresholds were used for gap filling on corresponding metabolic network.
Table 3:
Optimal parameters for MeGaFiller.
