If you are a drug discovery firm dealing with slow lead optimization — this project developed ML-based models for predicting solubility and toxicity that speed up the selection of viable candidates. This reduces the time spent on failed molecules.
Rapid Drug Discovery Pipeline for Broad-Spectrum Antiviral Therapeutics
Imagine having a master key that can unlock and stop many different types of viral locks. Instead of starting from scratch for every new virus, this work creates a fast-track system to find and refine medicines quickly. It uses AI and special fat-bubbles to deliver these medicines exactly where they are needed in the body.
What needed solving
Developing new antivirals is slow and risky, often failing due to poor solubility, toxicity, or delivery issues. There is a critical need for a rapid-response system to identify and validate drug candidates before a pandemic escalates.
What was built
An early-stage drug discovery pipeline including two ML models for solubility/toxicity and a specific liposomal delivery system (Formulation C).
Who needs this
Who can put this to work
If you are a delivery specialist dealing with unstable antiviral compounds — this project developed a specific liposomal formulation (Formulation C) that efficiently encapsulates these compounds. This improves the stability and delivery of the active ingredients.
If you are a CRO dealing with the need for rapid response to emerging outbreaks — this project established an early-stage drug discovery pipeline. This allows for the rapid identification of compounds against diseases like Zika or Dengue.
Quick answers
What is the cost or pricing for using these tools?
Based on available project data, no specific pricing or cost structures for the developed models or formulations are provided.
Can this be scaled to industrial production?
The project has identified an optimal liposomal formulation (Formulation C) and developed synthetic fusogenic liposomes, which are key steps toward industrial scale-up of drug delivery.
What are the IP and licensing options for the ML models?
Based on available project data, the ML models for solubility and hERG toxicity were developed using proprietary data from a pharmaceutical partner, suggesting specific IP arrangements are in place.
How does this integrate into existing drug pipelines?
The project establishes an early-stage pipeline that integrates molecular modeling, ADMETox profiling, and animal models to move from validated hits to pre-clinical candidates.
What is the timeline for clinical availability?
The project period runs from 2024-01-01 to 2028-06-30, with a specific goal to conduct a Phase 2a clinical trial for a Zika virus small molecule.
Who built it
The consortium is heavily weighted toward research and academic expertise, with 10 research organizations and 8 universities. However, it maintains a practical edge with 3 industry partners (including 2 SMEs), resulting in a 14% industry ratio. Led by Fraunhofer, the group leverages a broad European network across 8 countries to bridge the gap between lab discovery and clinical trials.
Contact Fraunhofer Gesellschaft for details on the antiviral pipeline and liposomal formulations.
Talk to the team behind this work.
Contact SciTransfer to explore licensing opportunities for the ML toxicity models or Formulation C.