Rasayan Labs, ICT Mumbai Launch AI Drug Discovery Platform

AI drug discovery is beginning to reshape the earliest stages of pharmaceutical R&D, moving beyond conventional compound screening toward computational molecule design, candidate optimization and synthesis planning.

The launch of Rasayan Lab’s AI-native platform at the Institute of Chemical Technology (ICT), Mumbai, brings this shift into focus for India’s pharma ecosystem.

The platform combines molecular design, retro synthesis, ADMET prediction and chemical intelligence in one environment.

More importantly, its collaboration with ICT has produced a reported three-week design-to-candidate timeline for a dual COX-2/mPGES-1 inhibitor program, pointing to how AI in drug discovery could compress parts of the early research cycle.

How is AI changing the way drug candidates are designed?

Traditional drug discovery involves multiple stages of hypothesis generation, compound selection, synthesis and experimental testing. Researchers may need to evaluate large numbers of molecules before identifying candidates with suitable biological and chemical characteristics.

AI-powered drug discovery introduces computational models earlier in this process. Instead of relying only on experimental screening, researchers can use computational chemistry and machine-learning models to identify, compare and optimize potential molecules before committing laboratory resources.

Rasayan Labs platform is designed around this approach, bringing together molecular analysis, generative capabilities, chemical intelligence and predictive tools. According to the company, the platform includes ADMET prediction across more than 1,400 properties, impurity prediction and a proprietary molecular editor.

The significance for pharmaceutical R&D is not simply that AI can generate more molecular structures. The larger opportunity is to improve the selection process-helping researchers determine which molecules warrant further investigation, what properties may require optimization and which candidates should progress toward laboratory testing.

This could make AI drug discovery in India increasingly relevant as Indian pharmaceutical companies look to strengthen capabilities beyond established strengths in generics, APIs and contract research.

Also Read: 10 AI-Powered Manufacturing Technologies Improving Pharma Production

Can retro synthesis connect AI drug design with actual synthesis?

A molecule that looks promising computationally still has to be made. This is where retro synthesis becomes important. Retro synthesis works backwards from a target molecule to identify potential chemical transformations and starting materials that could be used to produce it.

In an AI drug discovery platform, integrating retro synthesis with molecular design can therefore bring an important manufacturing question into the discovery process: Can this molecule actually be made?

Rasayan Labs says its platform incorporates AI-powered retro synthesis and route planning, supported by a database containing more than 13 million reactions and 126 million building blocks.

This creates a potential bridge between molecular design and drug synthesis. Rather than evaluating a candidate only on predicted biological activity or molecular properties, researchers can also consider possible synthetic routes and chemical starting points earlier in the workflow.

That makes retro synthesis more than another feature of computational drug discovery.

Computational synthesis prediction cannot replace laboratory chemistry or commercial-scale process development, but it can potentially help researchers identify practical routes earlier.

What has the Rasayan Labs-ICT Mumbai collaboration demonstrated?

The Rasayan Labs and ICT Mumbai collaboration provides a practical example of this computational-to-experimental model. The partnership was formalized through a memorandum of understanding focused on areas including protein target identification and therapeutics development.

Earlier work under the collaboration reportedly produced a three-week design-to-candidate timeline for dual COX-2/mPGES-1 inhibitors in an inflammation program. The program identified 23 candidates, with 19 advanced into synthesis, according to Rasayan Labs.

The platform therefore sits at the intersection of AI-based drug candidate development, medicinal chemistry and experimental research.

Also Read: How Agentic AI Is Revolutionizing Autonomous DMTA Loops?

Why could AI drug discovery matter for India's pharma R&D?

For India, the larger opportunity lies in building a stronger drug discovery pipeline alongside its established pharmaceutical manufacturing base.

AI can potentially reduce the number of molecules that need to be physically synthesized and tested during early research by improving computational prioritization. At the same time, tools such as retro synthesis can bring questions around chemical feasibility closer to the point of molecule design.

For Indian pharmaceutical companies, research institutions and emerging biotech firms, connecting these stages could strengthen domestic capabilities in pharmaceutical innovation and original drug research.

But software alone will not create this ecosystem. Successful AI in pharmaceutical research and development still depends on high-quality chemical data, medicinal chemists, experimental validation, biological testing and eventually process-development expertise.

AI drug discovery is moving from simply predicting which molecules might work toward helping researchers ask whether those molecules can be designed, synthesized and experimentally validated. For India's pharma sector, that convergence could become an important part of building a more integrated R&D capability.