Leveraging the power of 3D AI

A new state-of-the-art 3D Quantitative Structure-Activity Relationship (QSAR) software package that builds statistical models (CoMFA, CoMSIA and HyPhar) based on data obtained from experimental assays.


The Problem

Many molecular properties can be predicted based on available information extracted from chemical structures and experimental assays. However, most of the available and used methods neglect the 3D geometrical properties of the molecules, which play a critical role in understanding the interaction between a ligand and a receptor. These properties are even more important when considering bioactive conformations of the molecules derived from experiments.

Missing 3D geometry

Missing 3D

Limited ligand-receptor

Limited ligand-receptor information

Undesired properties



We developed a 3D Quantitative Structure-Activity Relationship (QSAR) software package that builds statistical models (CoMFA, CoMSIA and HyPhar) based on data obtained from experimental assays.

Our tool uses a unique and superior 3D representation of molecules based on electrostatic, steric and hydrophobic interaction fields derived from semi-empirical Quantum-Mechanics (QM) calculations. Such fields describe with high accuracy the factors that determine ligand / receptor interactions. PharmQSAR is peer-reviewed and validated. See our publications for more information.

Improved candidate molecules

Improved candidate

Better activity predictions

Better activity

Improved searches


All the benefits


  • Ligand preparation: 2D-3D molecular conversion, structure minimization and conformation, tautomer and stereoisomer generation of your dataset.
  • High quality parameters calculation:
    Partial charges (Gasteiger, Mulliken, Electrostatic (AM1/RM1)).
    Atomic-level LogP contributions (semi-empirical, (RM1) IEF/PMC-MST solvation models).
  • Common file formats supported: SDF, MOL2, SMILES and InChi.
  • Precise molecular alignment. Highly accurate field-based molecular alignment using electrostatic, steric and hydrophobic interaction fields. Molecular alignment is a critical step in 3D QSAR studies.
  • Generation of isocontour maps for visualization with PyMol or JMol.
  • Output statistical endpoints: R2, SD, CV, Spress.
  • Default setup but fully configurable for advanced users.


• Predicting molecular properties.
• Pharmacophore generation.
• Visualizing relevant areas for ligand-receptor interaction. in order to:
• Improve your candidate molecules in the Lead
Optimization phase.
• See which areas of the molecule should be modified
• Prioritize follow-up compounds based on predicted
• Predict other key molecular properties of new
• Understand which factors drive the activity of
your leads.
• Improve your virtual screening searches in ligand libraries.


QSAR software package workflow

PharmQSAR allows to automatically generate 3D QSAR/QSPR models using a library of compounds with known activities or properties and their molecular fields. The generated model is validated with an external data set of molecules. The trained model permits to predict relevant properties such as important ligand-receptor interactions and evaluate these properties on new chemical libraries. PharmQSAR generates also projections for an easy visualization of the calculated properties.


Choose what works best for you and start using PharmQSAR: the versatility of the command-line interface, or the connectivity and scalability of our API with other platforms and operating systems.


Run on your IT infrastructure

Scalable calculations

Updates and support included


Seamless integration and scalability

Python library interface

AWS integration




Torre R, 4a planta, Despatx A05, Parc Científic de Barcelona (PCB). C/ Baldiri Reixac 4-8 08028 Barcelona