Skip to main content

Table 2 Results from PCA models on the PDB and DrugBank dataset.

From: A chemogenomics view on protein-ligand spaces

Descriptor set

#descriptors

R 2 X

Q 2

#components

Protein

147

0.695

0.554

10

Ligand

35

0.798

0.729

4

Ligand + Protein

182

0.638

0.552

10

  1. This table contains the results obtained from a principal component analysis on the dataset, described by protein or ligand descriptors in isolation, and by the combination of protein and ligand descriptors. The number of descriptors (#descriptors) is displayed along with the fraction of explained (R2X) and predicted (Q2) variation captured by the components (#components).