Green Test Logo
This prediction engine is unique because it uses three independent models with varying degrees of ligand bias. The models are:

  • Docking. This is a structure-based model with zero ligand bias. A recent publication describes the successful implementation and design of this approach using human estrogen receptor alpha as an example [1].
  • 3D Chemical Fields. This is a ligand based model with moderate ligand bias. The excellent performance of this approach is described in a recent publication [2]
  • QSAR. A QSAR approach using Partial Least Squares has significant ligand bias.

1. Rueda, M., Totrov, M. & Abagyan, R. ALiBERO: Evolving a Team of Complementary Pocket Conformations Rather than a Single Leader. J Chem Inf Model 52, 2705–2714 (2012).

2. Kufareva, I., Chen, Y.-C., Ilatovskiy, A. V. & Abagyan, R. Compound activity prediction using models of binding pockets or ligand properties in 3D. Curr Top Med Chem (2012).