AI-assisted Drug Discovery platform:

Auri.AI

The drug discovery and development is a protracted, resource-intensive process involving substantial investments, often reaching millions of dollars. Unfortunately, nearly 90% of drug discovery programs face failure at various stages, primarily due to the lack of platforms capable of predicting success or failure in initial phases.

Recognizing this formidable challenge and aligning with Aurigene’s mission, our experts have developed an AI-assisted drug discovery platform Auri.AI. This is a one-of-a-kind platform created through the integration of advanced AI algorithms and computational/CADD approaches. It has the potential to reduce timelines and investment requirements in the drug discovery process by 40%. This unique amalgamation expedites candidate nomination, supporting the entirety of drug discovery projects from hit identification to the selection of preclinical candidates.

The digital platform features a meticulously curated compound database which is validated with structure normalization rules and is ever-expanding using both in-house and public domain information. The robust database, which stands today at 180 million compounds, and 1.6 million validated bioassay data points, serves as a robust search engine and provides training data for the AI algorithms. The database is searchable through structure & property / semantic queries for getting relevant information required for a specific purpose.

Platform also offers significant amount of scope for customization by bringing several AI models onto one screen. This platform is a one stop shop for a wide variety of top-tier predictive and generative AI models, as well as computational/CADD models. To date this platform consists of 11 generative AI, 4 predictive AI and 8 CADD/computational models. This modular and embedded design enables our scientific experts to cherry-pick the right set of models to be deployed for a specific application.

Being an end-to-end service provider, Aurigene is well equipped to quickly synthesize and run a variety of experimental assessments in both in-vitro and in-vivo . The results from the wet lab testing can be fed into our platform in real-time for training the AI models, enabling it to improve further. Creating this iterative-cycle, makes our platform unique and ensures that each new chemical design prediction made by the platform is increasingly accurate and reliable, and has the desired properties.

‘The proof of the pudding is in the eating’ – Early data using our AI platform shows a hit rate of 30% right in the first iteration of the DMTA cycle.

In conjunction with our synthetic chemistry and discovery biology services, this platform offers a comprehensive solution to clients eager to unveil and introduce their ground-breaking molecules to the market. It facilitates early detection of potential failures, paving the way for prompt reinitiation of candidate selection cycle. This approach significantly enhances the likelihood of successful drug discovery within a condensed timeframe and minimized costs.

Speak to our experts