Drug discovery takes too long. Traditional research and development can stretch for years, with most potential treatments failing in expensive late-stage trials. Persistent Systems thinks AI can change that.
The Indian tech company announced a partnership with NVIDIA to build AI tools that simulate drug molecules before researchers test them in real labs. The goal is to help pharmaceutical companies identify promising drug candidates faster and avoid costly dead ends.
How does it work?
Persistent built a system called GenMolVS (Generative Molecules and Virtual Screening) using NVIDIA’s BioNeMo platform. The tool works like this:
- AI models simulate how molecules behave and interact with biological targets
- The system screens thousands of potential drug compounds virtually
- Smart agents prioritize the most promising candidates for real lab testing
- Researchers get recommendations for which experiments to run next
The platform runs on NVIDIA’s AI infrastructure, including specialized processors designed for complex calculations. It uses large language models trained on chemical and biological data to predict how new molecules might work as medicines.
Why does it matter?
The pharmaceutical industry faces enormous pressure to develop new treatments faster. Current drug development costs billions of dollars and takes over a decade on average. Most experimental drugs fail.
Virtual screening could help companies:
- Test more potential drugs without expensive lab work
- Identify failures early, before costly human trials
- Focus lab resources on the most promising candidates
- Speed up the entire discovery process from months to days for initial screening
Persistent claims its AI system maintains scientific accuracy while dramatically reducing research timelines. The company says it can help pharmaceutical clients move from “months-long experiments to AI-driven discovery in days.”
The context
This partnership reflects a broader shift toward AI in drug discovery. Major pharmaceutical companies are increasingly using machine learning to identify new treatments and optimize existing ones.
NVIDIA has positioned itself as a key player in this space with its BioNeMo platform, which provides pre-trained AI models for biological research. The company’s graphics processors, originally designed for gaming, now power much of the AI revolution across industries.
Persistent Systems, based in India with over 26,500 employees worldwide, specializes in digital engineering services. The company has been expanding its AI capabilities and sees healthcare as a major growth area.
The collaboration also includes plans to train Persistent’s engineers on NVIDIA’s AI tools and expand certification programs. This could help the company build more sophisticated AI applications for pharmaceutical clients in the future.