The Application Note “AIDDISON: Empowering Drug Discovery with AI/ML and CADD Tools in a Secure, Web-Based SaaS Platform” from Rusinko et. al. was published in “Journal of Chemical Information and Modeling” 2024.

The landscape of drug discovery is rapidly evolving, thanks in large part to the integration of artificial intelligence (AI), machine learning (ML), and advanced cheminformatics tools. The publication highlights AIDDISON™ AI-Powered Drug Discovery, a secure, web-based platform that brings together AI/ML and computer-aided drug design (CADD) tools to accelerate the identification and optimization of new drug candidates. What sets AIDDISON™ apart is its intuitive interface, which lowers the barrier for scientists who may not be experts in computational methods, making cutting-edge drug discovery accessible to a wider range of researchers.

The workflow of AIDDISON™ begins with the generation or identification of thousands of viable molecules, leveraging a variety of sources such as similarity searches in public databases, 2D pharmacophore screening, and de novo molecular design using generative models. The platform then applies property-based filtering, molecular docking, and shape-based alignment to prioritize molecules with the highest probability of biological activity and optimal ADMET (absorption, distribution, metabolism, excretion, and toxicity) profiles. One of the key innovations is the seamless integration with SYNTHIA^© Retrosynthesis Software, which allows users to immediately assess the synthetic accessibility of promising molecules and identify necessary reagents for laboratory synthesis.

The power of this integrated approach was demonstrated in a case study focused on tankyrase inhibitors—a class of molecules with potential anticancer activity. Starting from a known inhibitor, the generative models of AIDDISON™ and virtual screening explored vast chemical space, producing a diverse set of candidate molecules. These were then filtered and docked to the target protein, and the most promising structures were sent to SYNTHIA^© for retrosynthetic analysis. This workflow not only accelerated the identification of novel, synthetically accessible leads but also enabled a more thorough exploration of chemical space than traditional methods alone.

For scientists and organizations looking to streamline their drug discovery pipelines, the combination of AIDDISON™ and SYNTHIA^© offers a robust, user-friendly solution. By bridging the gap between virtual molecular design and practical synthesis planning, these tools empower chemists to innovate faster and with greater confidence—ultimately bringing new therapies to market more efficiently.

Read the full article


AIDDISON™: Empowering Drug Discovery with AI/ML and CADD Tools in a Secure, Web-Based SaaS Platform
Andrew Rusinko, Mohammad Rezaei, Lukas Friedrich, Hans-Peter Buchstaller, Daniel Kuhn, and Ashwini Ghogare; Journal of Chemical Information and Modeling 2024 64 (1), 3-8
DOI: 10.1021/acs.jcim.3c01016