Skip to main content
Figure 3 | BMC Bioinformatics

Figure 3

From: Information-based methods for predicting gene function from systematic gene knock-downs

Figure 3

Network and subnetwork comparisons. Key is shown in the upper left-hand panel. A. AGREE links join genes with more shared present phenotypes. The x-axis indicates the link complexity, or number of shared present phenotypes. The y-axis indicates the frequency of occurrence of each link complexity bin. B. AGREE links have higher precision for every level of link complexity. The links of each network were binned by their complexity (x-axis). The y-axis indicates the precision of each bin for each network. Error bars indicate one standard deviation, assuming a binomial distribution. C. AGREE subnetworks are enriched for more phenotypes. The number of over-represented phenotypes present in the genes of each subnetwork (subnetwork complexity) was determined using the hypergeometric distribution (see Methods). The x-axis indicates subnetwork complexity. The y-axis indicates the frequency of the given subnetwork complexity in the ANET and UNET. D. A greater number of AGREE links are supported by other data types. The x-axis indicates the data type. The y-axis indicates the number of links in the ANET and UNET which are supported by that data type. Error bars indicate one standard deviation, assuming a binomial distribution.

Back to article page