About SuperPred

Here, we present SuperPred, which is a prediction webserver for ATC code and target predicition of compounds.
Predicting ATC codes or targets of small molecules and thus gaining information about the compounds offers assistance in the drug development process.
The webserver's ATC predicition as well as target prediction is based on a pipeline consisting of 2D, fragment and 3D similarity search.

The drug classification for a compound can be performed at the Drug Classification site. Target prediction for an input compound can be executed at the Target-Prediction site.

Statistics of the testset and the cross-validation results can be found here.

If you have any questions please see the FAQs or feel free to contact us!

Please cite:
Nickel J., Gohlke B.-O., Ehreman J., Banerjee P., Rong W.W., Goede A., Dunkel M. and Preissner R.
SuperPred: update on drug classification and target prediction.
Nucleic Acids Res 42(Web Server issue): W26-31. (2014)

ATC prediction

The ATC code prediction is based on a similarity search pipeline including 2D, fragment and 3D similarity. The users' query compounds will be screened against 2,600 known compounds having ATC codes. All three methods are applied for the calculation.


The Anatomical Therapeutic Chemical (ATC) classification system is used for the classification of drugs. It is published by the World Health Organization (WHO). The classification is based on therapeutic and chemical characteristics of the drugs. Each ATC code is divided into 5 levels:

  • 1. level: Anatomical main group
  • 2. level: Therapeutic main group
  • 3. level: Therapeutic/pharmacological subgroup
  • 4. level: Chemical/therapeutic/pharmacological subgroup
  • 5. level: Chemical substance
Substances or combination of substances in the 5th level refer to a single indication. Drugs having more than one indication belong to more than one ATC code. Aspirine for example has 3 ATC codes assigned.

Target prediction

The target prediction is based on the similarity distribution among the targets’ ligands. The distributions are utilized for estimating individual thresholds and probabilities for a specific target. By means of these individual thresholds and probabilities the input compound is screened against a database containing about 341,000 compounds, 1,800 targets and 665,000 compound-target interactions.