Markup | Genome Research

Table 7.

Characteristics of Gene Prediction Accuracy for the E. coliGenome with 4289 Annotated Genes

Models	AT cluster type	Sn	Sp	Average (Sn, Sp)	No. of correctly predicted genes	No. of predictions
GB		0.908	0.978	0.943	3895	3981
R		0.851	0.988	0.919	3651	3696
AT	2A	0.925 (0.923)	0.982 (0.983)	0.954	3968	4039
AT,T	2A	0.931 (0.929)	0.981 (0.982)	0.956	3995	4073
R,AT,T	2A	0.932	0.978	0.955	3999	4087
T	2A	0.835 (0.834)	0.991 (0.991)	0.913	3580	3614

AT	2B	0.916 (0.911)	0.985 (0.985)	0.951	3930	3991
AT,T	2B	0.922 (0.917)	0.983 (0.983)	0.953	3954	4023
R,AT,T	2B	0.922	0.983	0.953	3954	4023
T	2B	0.859 (0.857)	0.990 (0.990)	0.924	3683	3719

AT	2C	0.924 (0.924)	0.980 (0.981)	0.952	3964	4043
AT,T	2C	0.939 (0.938)	0.979 (0.979)	0.959	4027	4113
R,AT,T	2C	0.939	0.979	0.959	4027	4113
T	2C	0.848 (0.845)	0.990 (0.990)	0.919	3639	3674

AT	3A	0.927 (0.924)	0.982 (0.983)	0.955	3975	4048
AT,HT,T	3A	0.936	0.978	0.957	4014	4103
AT,T	3A	0.934 (0.930)	0.980 (0.981)	0.957	4004	4085
HT	3A	0.716 (0.715)	0.989 (0.991)	0.853	3070	3103
R,AT,T	3A	0.934	0.980	0.957	4004	4085
R,AT,T,HT	3A	0.936	0.978	0.957	4014	4103
T	3A	0.852 (0.850)	0.990 (0.990)	0.921	3653	3689

AT	3B	0.922 (0.916)	0.984 (0.984)	0.953	3956	4022
AT,HT,T	3B	0.931	0.980	0.956	3993	4076
AT,T	3B	0.929 (0.922)	0.982 (0.982)	0.956	3983	4058
HT	3B	0.713 (0.715)	0.990 (0.991)	0.851	3057	3089
R,AT,T	3B	0.929	0.982	0.956	3983	4058
R,AT,T,HT	3B	0.931	0.980	0.956	3993	4076
T	3B	0.872 (0.869)	0.989 (0.989)	0.930	3741	3781

AT	3C	0.926 (0.922)	0.980 (0.980)	0.953	3971	4052
AT,HT,T	3C	0.943	0.976	0.960	4046	4146
AT,T	3C	0.941 (0.938)	0.977 (0.978)	0.959	4035	4128
HT	3C	0.713 (0.717)	0.990 (0.992)	0.851	3057	3089
R,AT,T	3C	0.941	0.977	0.959	4035	4128
R,AT,T,HT	3C	0.943	0.976	0.960	4046	4146
T	3C	0.863 (0.860)	0.989 (0.990)	0.926	3701	3741

[i] The results obtained by using preclustering cross validation are shown in parentheses (italics). Boldface numbers show the maximum postclustering value of Avg(Sn, Sp) for a given species.