top of page

RNA Ligand Diffusion

Our two-part system addresses the critical need for specificity in RNA-targeted drug development. By leveraging cutting-edge AI and deep learning models, we create a robust solution that identifies binding sites and generates small molecule ligands with desired ADMET properties. This innovative approach promises to significantly advance RNA-targeted therapeutics, offering a compelling opportunity for investment and collaboration.

Introduction:

In the dynamic field of drug development, targeting RNA offers expansive opportunities. Currently, less than 700 proteins are drug targets, representing a mere 0.05% of the human genome. Utilizing RNA as a target vastly increases these possibilities. Our proposal aims to revolutionize RNA-targeted therapy by developing small molecule ligands that selectively bind to RNA targets, overcoming significant current challenges.

Challenges in RNA-Targeted Drug Discovery:

Difficulty in determining RNA structures using current in vitro methods.
Developing small ligands that bind to the found binding sites
Limited RNA structures in the Protein Data Bank.

Innovative Approach:

Detection of RNA Binding Sites: Leveraging deep learning models to identify appropriate binding sites on RNA, ensuring selectivity.
Development of Small Ligands: Creating ligands that specifically bind to these sites, overcoming limitations of existing databases and high fail rates.
Data Scarcity Solutions: Utilizing transfer learning, synthetic datasets, and combining multiple databases to enhance model training and effectiveness.

Value Proposition:

Enhanced Specificity and Selectivity: Our Software GeneSys is able to generate de novo RNA-targeted small molecules by using the whole scope of the chemical space with the help of generative AI and adapting to the newest findings.
Innovative AI Utilization: By adapting advanced deep learning models from protein targeting to RNA, which pushes the boundaries of drug discovery.
Cost and Time Efficiency: The AI-driven approach reduces reliance on costly and time-consuming wet lab experiments, accelerating the validation of drug candidates.

Our two-part system addresses the critical need for specificity in RNA-targeted drug development. By leveraging cutting-edge AI and deep learning models, we create a robust solution that identifies binding sites and generates small molecule ligands with desired ADMET properties. This innovative approach promises to significantly advance RNA-targeted therapeutics, offering a compelling opportunity for investment and collaboration.

We invite further discussion and exploration of our proposal, confident in its potential to transform RNA-targeted drug discovery and development.

bottom of page