Research

At the Volkamer Lab, we develop novel methods at the interface between structural bioinformatics and cheminformatics, mostly applied in the context of computer-aided drug design. Here, the individual projects are grouped into the following research lines. For more information click on the individual topics.

Research lines

Structure-based (binding site centric) approaches

• Binding site comparison
  • Targetome-wide binding site comparison
  • Kinome-wide binding site comparison
• Pharmacophore modeling
  • Static target-based pharmacophores
  • Dynamic target-based pharmacophores
• Kinase-focused methods
  • Kinase-fragment library
  • Novel Kinase Ligand Generation using Subpocket-Based Docking
  • Deep learning based ligand generation from kinase fragments
  • Kinodata-3D
  • more at OpenKinome
• MolDockLab
• DockM8 (advanced consensus docking workflow)

Machine learning based methods

• ML-based toxic endpoint prediction

  • KnowTox
  • Cytotoxicity
  • Morpheus

• Deep learning based virtual screening

• Data augmentation and deep learning for physico-chemical property prediction

• Molecular Property Representation and Optimization using Transformers

• Historical Virtual Control Groups

• Kinodata-3D

• more deep learning based work at KinFragML

OpenKinome

• KinoML
• MIAME
• More kinase related work at KiSSim and KinFragLib