Platform Technology


Executive Summary

An Ensemble of Hybrid Techniques.

In-silico Drug Design Platform

PharmCADD combats one of the grandest challenges in the pharmaceutical R&D – It’s innovative AI platform PharmulatorTM can predict the intricate 3D structures of proteins based solely on their amino acid sequences within a few minutes.

To circumvent the problems arising from the conventional rigid protein-ligand docking models, PharmulatorTM utilizes in-vivo environmental features and complex physicochemical properties of molecules as its input database for AI training, enabling the predictions of dynamic “evolution” and functional free energy of target protein-ligand bindings.

Protein 3D Structure

All proteins are made out of 20 amino acids, but combined in different ways. The way these 20 amino acids are arranged dictates the folding of the protein into its unique final shape.

Our Research Focus on Membrane Proteins

Membrane Proteins are primary targets of drugs.

MP (Membrane Protein) : 54% of FDA approved drug binding targets
GPCR (G-Protein Coupled Receptor) : 34% of approved drug binding targets

Protein Structure Prediction Using Neural Network

Pharmulator can predict the 3D structures of proteins based solely on amino acid sequences.

Structure & Dynamics & Function

Contribution of Dynamics

Molecular Dynamics Supporting System (automation is ongoing)

CHEMBL Diversity Dataset

Small Molecule (Parameter) : Quantum DFT + Training

Small Molecule (Parameter) : Quantum DFT + Training


hERG-related drug toxicity

Formulator: Early stage prediction of ADMETox parameters

– Ex: hERG
– potassium ion channel of the human ether-a-go-go-related gene (hERG)
– Blockage leads to long QT syndrome (LQTS), which is a severe life-threatening cardiac side effect.
– Predict drug-induced hERG-related cardiotoxicity could facilitate drug discovery by filtering out toxic drug candidates.

hERG Machine Learning model


hERG-related drug toxicity

WEB User-Friendly Platform
The influence of each atom or functional group on these properties can be highlighted and combined with visualization


hERG-related drug toxicity