Chemistry42: An AI-Driven Platform for Molecular Design and Optimization

    Yan A. Ivanenkov, Daniil Polykovskiy, Dmitry S. Bezrukov, Bogdan Zagribelnyy, Vladimir Aladinskiy, Petrina Kamya, Alex Aliper, Feng Ren, Alex Zhavoronkov
    TLDR Chemistry42 effectively creates and optimizes new molecules for drug discovery.
    Chemistry42 was an AI-driven platform developed by Insilico Medicine to enhance molecular design and optimization, integrating over 40 generative models to produce novel molecular structures with desired properties. It was utilized by over 20 pharmaceutical companies and involved more than 30 internal programs. The platform's workflow included data upload, generation of novel structures, and analysis, employing models like generative autoencoders and adversarial networks. Chemistry42 facilitated rapid drug discovery, as demonstrated by the discovery of DDR1 kinase inhibitors in 21 days, and supported workflows for ligand- or structure-based drug design. Its interoperability with bioinformatics services and user-friendly interface aided decision-making in drug design, showcasing AI's growing role in pharmaceutical research.
    Discuss this study in the Community →

    Research cited in this study

    1 / 1 results