Skip to main content
Fig. 4 | BMC Bioinformatics

Fig. 4

From: LANDMark: an ensemble approach to the supervised selection of biomarkers in high-throughput sequencing data

Fig. 4

Principal Coordinate Analysis projections of test data can be used to assess model fit. Proximity matrices extracted from the Extremely Randomized Trees, Random Forest, LANDMark (Oracle), and LANDMark (No Oracle) models trained on the Wisconsin Breast Cancer dataset. Higher amounts of explained variance along the first principal component, relative to other models, reflect the ability of a model to identify a set of simple decision pathways capable of classifying samples. Higher explained variance along additional components (relative to the other models) suggest the presence of complex decision pathways and overfitting

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us