Fig. 2From: Evaluation and improvement of the regulatory inference for large co-expression networks with limited sample sizeComparison of pAUPR values for different methods and different network structures. Each bar in the plots represents mean of pAUPR values from the top 1000 edge predictions. Error bars represent standard error. The differences of pAUPR values between different methods were determined using a Student t-test in pairs between RLowPC and the other eight methods. P-values are shown on the top of the bars if it is less than 0.05Back to article page