A novel tool to predict small molecule binding into proteins


PANTHER - effective virtual screening

  • Rapid

  • Evaluate tens of millions of small molecules into selected protein target in 24 hours (in our computing system).
  • Accurate

  • PANTHER identifies active molecules typically with higher accuracy than any other published method.
  • Considers the protein

  • Structure


    Physico-chemical properties

    Binding site solvation

Panther logo

"If there is a structure available for a potential target protein of a chemical, methods that we have developed are applicable."

Olli T. Pentikäinen

One of the key advantages of PANTHER is that it enables the user to optimize the ligand binding into selected target by producing binding predictions into wide panel of optimized PANTHER-models that represent different proteins, and pharmacological response.

PANTHER can be applied into wide variety of problems, for example:

1. Reduce the failures at the clinical phases of drug development.

2. Estimate the environmental toxicity of given chemical.

3. Re-scoring of docking results (see Kurkinen et al., 2018)




The Panther code (python) is available under GPL license, which means it is freely available (== $0). Panther-0.18.21 can be downloaded from here. It requires an e-mail address and name, which are collected only to get an idea of how many users there are.



If you are interested to collaborate with us to identify novel small molecules for your target protein, or, to understand the risks that your chemical may cause with it's binding into target proteins that are not intended, you are welcome to ask for our view on the matter.

For example, in the drug discovery context our strategy comprises of the following stages. At the first stage, we check that there is appropriate data available - if yes - in the second stage we will screen both commercially available compound libraries and our own molecular database. At the third stage, we can both optimize the ligands and synthesize them.

Our modeling service is available in commercial context as well. Pricing is available on request.

Please, do not hesitate to ask!




If using Panther, remember proper citation:
Niinivehmas SP, Salokas K, Lätti S, Raunio H, Pentikäinen OT. Ultrafast protein structure-based virtual screening with Panther. Journal of Computer-aided Molecular Discovery, 2015, 29(10):989-1006.


We have published a practical perspective on usage of PANTHER in virtual screening. Jointly with this publication, we published a major update of Panther-code. Furthermore, useful scripts were published with the paper.
Dowload pdf
Supplementary files (contains an example case for virtual screening by using free/academic software tools)

You may find other papers from us that refer into usage of Panther, so check out our list of publications.

Please, do not hesitate to ask!

Olli Pentikäinen E-mail: