Skip to main content

Advertisement

Figure 3 | BMC Systems Biology

Figure 3

From: ENNET: inferring large gene regulatory networks from expression data using gradient boosting

Figure 3

The analysis of the sampling rates s s and s f for DREAM 4 size 100 challenge: the Overall Score and a loss. The analysis of the sampling rates s s and s f for the DREAM 4 size 100 challenge. A: For each set of parameters (s s , s f , M, ν)  {0.1,0.3,0.5,0.7,1} × {0.1,0.3,0.5,0.7,1} × {5000} × {0.001} we analyzed an average 5-fold cross-validated loss over all the observations (across all gene selection problems) from all 5 networks. The minimal average loss was achieved for high values of s s  = 1 and low values of s f  = 0.3. B: We also compared the measure based on an average loss with the original Overall Score, as proposed by the authors of the DREAM challenge. The results were consistent across the two measures, i.e., the parameters that gave low average loss also led to accurate network predictions (a high Overall Score).

Back to article page