A peer-reviewed study found that the company's new technology enables faster dataset construction, further accelerating Avicenna's life-saving drug development timeline.
Durham, North Carolina, May 8, 2024–(BUSINESS WIRE)–Avicenna Biosciences today introduced enhancements to its machine learning (ML) technology platform to enhance medicinal chemistry and accelerate clinical-stage drug discovery. The company has raised $14.5 million in funding to date, with DCVC Bio leading a 2022 seed round, and published a paper this month in the peer-reviewed Journal of Chemical Information and Modeling. The paper, co-authored by Schrödinger and researchers at Microsoft Research AI4Science, combines Schrödinger's physics-based approach with Avicenna's new ML approach to more quickly move through the lead-to-drug optimization phase of small molecule drug discovery. outlines how you can be more successful at lower costs. This is especially true when it comes to the efficacy and selectivity of engineering against potential biological targets.
“While we are used to hearing about scientific success stories, the countless failures that occur along the way are often overlooked.Failures are especially prevalent in medicinal chemistry.To bridge the complex gaps, “Many clinical endeavors can cost hundreds of millions of dollars,” said Avicenna's co-founder and chief scientific officer. said Dr. Thomas Kaiser. “Avicenna is applying new ML methodologies to navigate and design around the unknowns in medicinal chemistry. Our methods allow drug design teams to learn from years of failures and We are making the most critical step in drug discovery much faster and more cost-effective. Become. ”
Application and results of technology
Avicenna is leveraging its technology to develop unique treatment programs, initially focused on neurodegenerative diseases. For example, Rho kinase (ROCK) inhibitors have shown potential in neurodegenerative and metabolic diseases. However, designing orally administered central nervous system-penetrating ROCK inhibitors has proven extremely difficult. Fasudil, a promising ROCK inhibitor currently in phase 2 clinical trials, must be administered intravenously twice a day and cannot be used for diseases such as chronic kidney disease or neurodegeneration. To overcome these challenges, Avicenna launched his own ROCK inhibitor program. By using new ML technology to identify drug-like compounds with desirable pharmacokinetic properties, the company has already achieved:
-
Faster timeline: 9 months from idea in vivo proof of concept
-
Reduce costs: $220,000 from conception to initiation of research enabling new drug investigation (IND)
-
Better treatment: two Discovery of development candidates while synthesizing only 11 types of compounds
Meaning of research paper
“Developing new drugs and safely delivering them to people who need them is incredibly difficult and risk-filled. Avicenna is committed to helping identify molecules with ideal drug-like properties. “We are working to make drug discovery easier and faster by creating new algorithms,” he said. John Hamer, his managing partner at DCVC Bio, said: “Avicenna's collaboration with Schrödinger and Microsoft shows that physics-based ML extensions require only tens of molecules, rather than thousands, to optimize small molecules for new drug targets. We couldn't be more excited to support this potential game in transforming biotech companies as they scale up to their next stage of growth.”
A newly published paper by Avicenna, Schrödinger, and Microsoft, titled “FEP-Augmentation as a means to solve the data scarcity problem of machine learning in chemical biology,” describes how Schrödinger's free energy perturbation technique, FEP+, Using physics-based techniques, virtual data can be used to augment sparse datasets commonly found in early medicinal chemistry optimization. These expanded datasets can be used for ML training that was previously not possible without the disclosed approach. This expansion is beneficial for ML training, as is the effort and expense of creating and testing the required compounds. Ultimately, this will enable early-stage discovery teams to access ML and rapidly query millions of lead-like compounds to identify promising leads for drug development. This paper outlines important mechanistic considerations for augmenting such datasets and lays the foundation for implementation of 10-20 related compounds with 3D structures co-resolved with a small set of ligands. We demonstrate that an initial series of can serve as an accelerated dataset. The foundation of lead optimization.
This study shows that the combination of FEP+ and ML can accelerate the hit-to-lead stage of drug discovery, with the potential to produce significant impacts, including:
Team and partnership opportunities
The Avicenna team's combined background in mathematics, chemistry, and medicine brings a unique and comprehensive perspective to the core areas of medicinal chemistry. Co-founders Drs. Kaiser and Peter Berger met while working in the Liotta Research Group at Emory University. The group is well known for its successful drug development across virology, oncology and neurology, producing more than 20 FDA-approved therapeutics. Dr. Kaiser led the Liotta antiviral drug group as a synthetic organic chemist, and Dr. Berger led its computational group as a structural bioinformatician.
In addition to developing proprietary therapeutic programs, Avicenna partners with clinical-stage biotech startups and large pharmaceutical companies to optimize drug discovery campaigns, maximize risk-return profiles, and integrate partners' existing chemical workflows. easily adapted to. For more information about Avcenna and its technology platform, or to inquire about partnership opportunities, please visit www.avicenna-bio.com.
About Avicenna Bioscience
Founded in 2020, Avicenna is on a mission to solve intractable drug design challenges that have previously hindered the development of drug candidates. The company's machine learning-powered medicinal chemistry platform makes lead-to-drug optimization faster, cheaper, and more successful, turning suboptimal clinical candidates into life-saving drugs. Avicenna is backed by his DCVC Bio, and Entrepreneur-in-Residence Christopher S. Meldrum joins the company as President and CEO. For more information, visit www.avicenna-bio.com and follow the company on LinkedIn.
View source version on businesswire.com. https://www.businesswire.com/news/home/20240508697108/ja/
contact address
media contact
kelly walker
kerry@walkercomms.com
