Skip to main content
Figure 4 | BMC Systems Biology

Figure 4

From: Detection of driver metabolites in the human liver metabolic network using structural controllability analysis

Figure 4

The fractions of the metabolites with different roles based on different frequency threshold. A) and B) respectively show the results which are based on the modules detected by the simulated annealing algorithm and the fast greedy algorithm. Each point connected by dotted lines is the fraction of the metabolites with a specific role among the set of driver metabolites whose frequency f dt  ≤ f d  < 1, while each solid line means the fraction of metabolites with each role among the HLMN. In the HLMN, most metabolites are of roles R1 and R2. With the frequency threshold f dt increasing, the fraction of R1 metabolites among the set of metabolites with f dt  ≤ f d  < 1 decreases while the fraction of R2 metabolites increases. The fractions of metabolites with different roles fluctuate when f dt  ≥ 0.7 due to the small size of the set of metabolites with f dt  ≤ f d  < 1. When f dt  < 0.7, the difference between the fractions of R1 metabolites and R2 metabolites is the biggest at about f dt  = 0.6. Thus, we choose the threshold f dt  = 0.6 to differentiate the high-frequency driver metabolites from the low-frequency driver metabolites, and the high-frequency driver metabolites tend to be of role R2.

Back to article page