Avicenna Biosciences, headquartered in Durham, N.C., has announced advancements in its machine learning (ML) technology platform designed to enhance medicinal chemistry and accelerate clinical-stage drug discovery. The company, which has secured $14.5 million in funding led by DCVC Bio, recently published research in the Journal of Chemical Information and Modeling. This study, co-authored with Schrödinger and Microsoft Research AI4Science, highlights how integrating Schrödinger’s physics-based methods with Avicenna’s ML techniques can expedite and economize the lead-to-drug optimization phase, especially in enhancing potency and selectivity against biological targets.
Dr. Thomas Kaiser, Avicenna’s co-founder and Chief Scientific Officer, emphasized the frequent failures in medicinal chemistry and the substantial financial investments required. He noted that Avicenna’s novel ML methods address these challenges by allowing drug design teams to learn from past failures and optimize drug targets more efficiently. This approach aims to make the critical, final phase of drug design significantly quicker and more cost-effective.
Avicenna is initially applying its technology to its therapeutic programs, focusing on neurodegenerative diseases. A notable example is their work on Rho kinase (ROCK) inhibitors, which show promise in treating neurodegeneration and metabolic diseases. Current ROCK inhibitors, like Fasudil, require intravenous dosing, limiting their practicality for chronic conditions. To address this, Avicenna launched its ROCK Inhibitor Program, using its ML technology to identify compounds with desired pharmacokinetic properties. This initiative has already yielded impressive results:
- Accelerated timelines: Achieving in vivo proof of concept in just 9 months
- Cost efficiency: Spending only $220,000 from concept to the initiation of Investigational New Drug (IND)-enabling studies
- Improved therapeutics: Discovering two development candidates after synthesizing only 11 compounds
Dr. John Hamer, Managing Partner at DCVC Bio, remarked on the collaborative research with Schrödinger and Microsoft, underscoring how the fusion of physics-based augmentation with ML requires fewer molecules to optimize small molecules against new drug targets. This approach significantly reduces the number of molecules needed from thousands to just tens, which could revolutionize drug discovery.
The research paper, titled “FEP-Augmentation as a Means to Solve Data Paucity Problems for Machine Learning in Chemical Biology,” discussed using Schrödinger’s free energy perturbation technology (FEP+) to generate virtual data, augmenting the sparse datasets typical in early drug optimization. This augmented data is valuable for ML training, enabling early-stage discovery teams to quickly query millions of compounds and identify promising leads for drug development. The paper emphasizes that combining FEP+ with ML can significantly shorten the hit-to-lead optimization time and reduce synthetic efforts.
Avicenna’s team combines expertise in mathematics, chemistry, and medicine. Co-founders Dr. Kaiser and Dr. Pieter Burger met at Emory University’s Liotta Research Group, known for developing over 20 FDA-approved therapeutics. Dr. Kaiser led the antivirals group as a synthetic organic chemist, while Dr. Burger headed the computational group as a structural bioinformaticist.
Besides developing its therapeutic programs, Avicenna collaborates with biotech startups and major pharmaceutical companies, optimizing their drug discovery processes and seamlessly integrating into existing workflows. Founded in 2020, Avicenna aims to overcome drug design challenges, making the lead-to-drug optimization process faster, cheaper, and more successful. Backed by DCVC Bio, the company is led by President and CEO Christopher S. Meldrum.
How to obtain the latest research advancements in the field of biopharmaceuticals?
In the Synapse database, you can keep abreast of the latest research and development advances in drugs, targets, indications, organizations, etc., anywhere and anytime, on a daily or weekly basis. Click on the image below to embark on a brand new journey of drug discovery!