Skip to main content
Figure 2 | BMC Bioinformatics

Figure 2

From: How accurate and statistically robust are catalytic site predictions based on closeness centrality?

Figure 2

(a) Probability density function of the closeness centrality scores for all catalytic (red) and all noncatalytic (blue) residues. The distributions are scaled such that the integral of each is equal to one. The solid line trend line is simply a running average meant to help guide the eye. The average values for the two distributions are 0.16 and 0.19, which is a statistically significant distinction (for a two-sample t-test: t = 2.0; p = 1.6E-73; sample size = 7,373). The overlap of the two distributions across the histogram of sampled data is 59.5%. (b) Probability density functions for all catalytic sites (red line) compared to the distributions of all at three different solvent accessibility levels. Each distribution is scaled such that its integral is one. Statistics describing the distribution comparisons are provided in Table 1.

Back to article page