May 8, 2024

ANYO Labs Secures SEK 1.46 Million Vinnova Grant to Accelerate AI-Powered Drug Discovery with de novo Molecule Generation

ANYO Labs is proud to announce that we have been awarded a SEK 1.46 million grant from Vinnova, Sweden's innovation agency, to advance our de novo (generative) capabilities. Through this grant, we aim to establish robust verification for our generative solution for virtual screenings as well as finding novel molecules for best-in-class therapeutics.

Addressing the Challenges of Traditional Drug Discovery

Traditional drug discovery is often plagued by high attrition rates, soaring costs and lengthy cycles. These limitations are often exacerbated by conventional methods like high-throughput screening (HTS) and traditional computational tools. These methods can struggle with accuracy and often rely on known chemical space, hindering the exploration of truly novel therapeutic options.

AI has emerged as a transformative force in drug discovery, offering the potential to significantly expedite the development process, ultimately bringing life-saving drugs to clinical trials more rapidly. In recent years, de novo drug discovery has become a promising avenue for uncovering best-in-class and entirely novel therapeutics. However, current computational limitations, coupled with the high costs associated with developing such molecules, have rendered the vast expanse of uncharted chemical space virtually inaccessible as well as proving that many AI algorithms often do not filter for activity, selectivity, synthesizability, and can create non-drug-like molecules (Gao & Coley, 2020).

It is rare to have the budget for true de novo validation and often, AI companies limit their AI generated molecules to befiltered by medicinal chemists to limit synthesis costs, which are magnitudes times higher than on-demand compounds. Multiple times, AI companies must incorporate highly manual medicinal and synthetic chemistry help to cut the high costs of generative AI compounds. Recently, Jang et. Al (2022) as an example chose only one compound to be synthesized out of 10,416 virtual hits and and Chenthamarakshan, et al. (2023) selected 4 out of 875,000 from their AI generated molecules.

iGen

ANYO Labs is at the forefront of leveraging AI to reshape drug discovery, particularly in the realm of de novo drug discovery through its i-TripleD platform. This innovative platform utilizes unique, structurally agnostic, cutting-edge AI/ML techniques, to identify and generate novel drug candidates that would be inaccessible through traditional methods.

The core strengths of i-TripleD lie in its ability to rapidly explore chemical space: The platform efficiently navigates vast, uncharted chemical territories at unprecedented speeds, through algorithms focused on the diversity of compound selection with the capacity to on average generate millions of hits in a day’s time. Most importantly, i-TripleD prioritizes candidates based on novelty, favorable chemical properties, and synthetic feasibility automatically through the proprietary iGen module.

Project details

With this grant, ANYO aims to undertake a rigorous, minimally assisted validation of our de novo methods, something rarely done at this scale. The key areas of focus will be:

  • Efficiency in de novo Drug Screening and Generation: We aim to establish new benchmarks for speed and efficiency in the generation of novel drug candidates.
  • Predicting Potency against Specific Targets: i-TripleD will be honed to predict the potency of generated compounds against targeted protein structures.
  • Accurate ADMET Property Estimation: The project will focus on validating the platform's ability to estimate and validate crucial ADMET (absorption, distribution, metabolism, excretion, and toxicity) properties for identified drug candidates in the first hit screenings, creating much more lead-like initial hits.

Outcomes

If successful this project will act as a testement of un-biased generative AI´s potential in drug discovery at scale for the industry. Our focus on efficiency and speed in generating these hits ensures a rapid transition from concept to pre-clinical development, providing a crucial advantage in the competitive pharmaceutical market. These lead-like hits not only offer stronger patent protection and a first-mover advantage but also hold the potential to address unmet medical needs, driving transformative innovation within the industry. As we progress through the rapid hit identification campaign, the team will carefully asses the total costs per hit and assess the quality of hits produced by iGen. This data will inform the development of a robust internal drug discovery pipeline to further validate our tools and methods.

A Future of Faster, More Efficient Drug Discovery

We are deeply grateful to Vinnova for their unwavering support and investment in our vision to revolutionize drug discovery. With this grant, we are confident in our ability to propel ANYO Labs to its full potential, paving the way for a future where drug discovery is faster, more efficient, and more accessible than ever before.

Stay tuned for further updates as we embark on navigating the uncharted territories of drug discovery.

  • Chenthamarakshan, V., Hoffman, S. C., Owen, C. D., Lukacik,P., Strain-Damerell, C., Fearon, D., ... & Das, P. (2023). Accelerating drug target inhibitor discovery with a deep generative foundation model. Science Advances, 9(25), eadg7865.
  • Gao, W., & Coley, C. W. (2020). The synthesizability of molecules proposed by generative models. Journal of chemical information and modeling, 60(12), 5714-5723.
  • Jang, S. H., Sivakumar, D., Mudedla, S. K.,Choi, J., Lee, S., Jeon, M., ... &Wu, S. (2022). PCW-A1001, AI-assisted de novo design approach to design a selective inhibitor for FLT-3 (D835Y) in acute myeloid leukemia. Frontiers in Molecular Biosciences, 9, 1072028.
original post