Project description
Machine learning for fast and accurate drug design
Drug discovery is a laborious and time-consuming process that needs to address multiple parameters, including the activity, solubility, stability and toxicity of prospective compounds. Although various computational approaches have been utilised to facilitate the design of new drugs, they suffer from speed and accuracy limitations. Funded by the European Innovation Council, the CelerisTx - Celeris One Platform project aims to develop a platform using machine learning to predict biomolecular interactions. Researchers plan to use this tool to design molecules that target protein-protein interactions and initiate the process of targeted protein degradation. This approach can be successfully employed to remove pathogenic proteins such as those associated with Alzeimer’s disease.
Objective
Celeris Therapeutics is a deep learning company that uses innovative, in-silico methods such as geometric deep learning and graph neural networks to degrade currently undruggable targets (pathogenic proteins). The platform shall make a broad impact by addressing currently incureable diseases such as Alzheimer's, Parkinson's, and different types of cancer like breast and prostate cancer.
Celeris Therapeutics' technical solution is the web application (orchestration platform) Celeris One.
It consists of three modules: Hades (target ID), Xanthos (predicting biomolecular interactions and ligand design), Hephaistos (automated synthesize and validate).
The addressed market is in the early-stage drug discovery and users are pharmaceutical and biotech companies with a focus on Targeted Protein Degradation.
The targeted customers are med. and comp. chemists, that currently rely on docking, which is computationally intensive, slow and inaccurate compared to CelerisTx deep learning methods.
Fields of science
- medical and health sciencesbasic medicinepharmacology and pharmacydrug discovery
- medical and health sciencesbasic medicineneurologydementiaalzheimer
- natural sciencesbiological sciencesbiochemistrybiomoleculesproteins
- natural sciencescomputer and information sciencesartificial intelligencemachine learningdeep learning
- natural sciencescomputer and information sciencesartificial intelligencecomputational intelligence
Programme(s)
- HORIZON.3.1 - The European Innovation Council (EIC) Main Programme
Funding Scheme
EIC-ACC - EIC-ACCCoordinator
8042 GRAZ
Austria
The organization defined itself as SME (small and medium-sized enterprise) at the time the Grant Agreement was signed.