Skip to main content
Figure 5 | BMC Bioinformatics

Figure 5

From: DART: Denoising Algorithm based on Relevance network Topology improves molecular pathway activity inference

Figure 5

Comparison of DART with CORG. Comparison of (unsupervied) DART with (supervised) CORG for the three different perturbation signatures: A) ERBB2 in ER- breast cancer, B) MYC in ER+ breast cancer, and C) TP53 in lung cancer/normal tissue. In the case of ERBB2 and MYC, Wang was used as the discovery set to learn the associated pruned relevance networks, while for TP53, the Landi set was used. The other cohorts are the test sets. y-axis labels the -log10(P-value), reflecting the degree of difference in predicted pathway activity levels between samples with and without the perturbation.

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us