Table 6.

Prediction on simulated missing GO annotations under BMA in MF sub-ontology

q		MicroAvgF1	MacroAvgF1	1-RankLoss	Fmax	RAccuracy
1	H →H	91.71 ±0.02	82.31 ±0.22	90.98 ±0.13	91.37 ±0.02	10.56 ±0.21
	M+H →H	93.70 ±0.02	82.33 ±0.21	96.91 ±0.03	93.87 ±0.02	32.30 ±0.26
	M →M	92.01 ±0.10	80.50 ±0.41	93.04 ±0.02	92.71 ±0.10	7.85 ±1.15
	H+M →M	93.09 ±0.02	80.60 ±0.40	96.78 ±0.03	93.27 ±0.02	19.59 ±0.21
2	H →H	80.69 ±0.05	57.31 ±0.41	80.40 ±0.17	80.04 ±0.05	25.12 ±0.21
	M+H →H	83.56 ±0.03	57.81 ±0.44	89.24 ±0.04	85.04 ±0.03	36.25 ±0.11
	M →M	79.02 ±0.01	54.82 ±0.41	78.43 ±0.02	79.26 ±0.01	15.78 ±0.05
	H+M →M	83.03 ±0.03	55.31 ±0.44	87.89 ±0.04	82.83 ±0.03	31.86 ±0.11
3	H →H	70.09 ±0.03	40.45 ±0.42	68.16 ±0.03	70.32 ±0.03	23.70 ±0.08
	M +H→H	74.76 ±0.04	41.37 ±0.39	80.44 ±0.04	76.58 ±0.04	35.62 ±0.11
	M →M	68.89 ±0.08	39.27 ±0.25	65.40 ±0.10	68.25 ±0.08	20.59 ±0.21
	H+M →M	73.68 ±0.04	40.18 ±0.33	76.63 ±0.07	72.99 ±0.04	32.81 ±0.09

q is the number of simulated missing annotations of a gene. H →H directly uses GO annotations of Human to predict annotations of Human genes. M+H →H uses GO annotations of genes from Mouse and Human to predict annotations of Human genes. M →M and H+M →M follow the similar protocol, but make prediction for Mouse genes. The data in boldface is the statistically significant best among these comparing methods for a particular target species, and the significance is checked by paired t-test at 95% confidence intervals