Skip to main content

Table 2 Prediction results by subtracting each feature using the 10-fold cross-validation on the training set

From: Computational identification of deleterious synonymous variants in human genomes using a feature-based approach

Feature

Recall

Precision

F-measure (β = 1)

AUC

All features

0.700

0.820

0.755

0.851

No SR-

0.697

0.823

0.755

0.851

No MES

0.667

0.810

0.731

0.829

No MDE

0.677

0.820

0.742

0.834

No DVE

0.697

0.816

0.752

0.846

No ese-dens

0.693

0.835

0.757

0.845

No dPSIZ

0.700

0.813

0.752

0.850

No verPhyloP

0.697

0.768

0.731

0.829

No TFBS

0.690

0.818

0.749

0.851

No DSP

0.683

0.837

0.752

0.845

No TE

0.673

0.798

0.731

0.829

  1. The highest values are highlighted in bold