Table 2 

Area Under the Curve scores for PSWM data sets 

Data set 
A.U.C 


3_08_1506 
79% 
3_07_551 
77% 
3_06_116 
76% 
4_08_1226 
76% 
4_07_447 
78% 
4_06_102 
77% 
5_08_778 
77% 
5_07_280 
77% 
5_06_62 
75% 
6_08_544 
76% 
6_07_202 
76% 
6_06_62 
76% 
Li_849 
76% 
LiFlank_849 
80% 


The AUC score is shown for each of the 14 PSWM sets when predicting operons using the positive and negative examples of operon members of E. coli. Data set nomenclature x_y_z refers to: x, Poisson distribution dimer significance threshold (log_{10 }P >) or Data set, y, Clustering threshold, and z, the number of clusters/PSWMs in data set (see also methods and text). 

Laing et al. BMC Genomics 2008 9:79 doi:10.1186/14712164979 