Gene regulatory network inference and validation using relative change ratio analysis and time-delayed dynamic Bayesian network

EURASIP Journal on Bioinformatics and Systems Biology

Table 1 Assessment metrics for the first set of E. coli and yeast networks inferred using our approach

Metrics	Ecoli1_10	Yeast1_10	Ecoli1_50	Yeast1_50	Ecoli1_100	Yeast1_100
AUPR	5.43E − 01	7.71E − 01	6.71E − 01	4.86E − 01	1.45E − 02	1.55E − 02
AUROC	7.94E − 01	9.44E − 01	8.62E − 01	8.35E − 01	5.21E − 01	4.61E − 01
P _AUPR	1.34E − 04	2.09E − 06	8.57E − 55	3.91E − 39	2.27E − 01	8.91E − 01
P _AUROC	5.47E − 04	1.29E − 06	3.19E − 20	4.64E − 18	2.02E − 01	9.60E − 01
Overall AUPR	1.09E − 04		2.54E − 46		4.83E − 03
Overall AUROC	2.10E − 04		8.19E − 18		2.13E − 02

The network name consists of two parts: organism name and network set number (i.e., Ecoli1 or Yeast1) followed by network size (10, 50, or 100 genes). The two parts are separated by ‘_’.