CALICI
AI-Powered Drug Discovery Designed for Real-World Science

CALICI began with a simple yet transformative vision to make artificial intelligence a practical, everyday tool for drug discovery. Founded in 2018, when AI was rapidly advancing, the company’s origins trace back to research at Baylor College of Medicine in Houston, Texas. While working on antiviral treatments for norovirus, the team tested hundreds of compounds using conventional in silico tools without success.

A discussion with co-founder and AI specialist Daniel led to a new idea of training AI on the team’s past results to cut down the number of trials. Yet existing AI tools required complex coding skills, creating a barrier for many biologists. Calici set out to change that by developing a fully web-based platform that brought advanced AI directly to the laboratory scientist.

The result is Pharmaco-Net^®, an online application that looks like a standard webpage but delivers sophisticated drug discovery capabilities. Guided by an AI agent called MediBot, users can upload protein structures, identify active sites, screen massive compound libraries, predict ADME/Tox properties, and refine molecular designs without requiring IT support.

“Our approach makes AI drug discovery practical and immediate, turning digital predictions into real compounds ready for the lab,” says Choi Jae-Mun, CEO.

Accessible, Actionable, and Built for Speed

Many AI drug discovery companies run their predictive models internally and provide results to clients. Calici takes a different approach by offering its AI tools directly to researchers. This allows scientists to interact with the technology in real time, making the process more collaborative and transparent.

The company also addresses a major industry challenge by turning AI-generated molecules into real, testable compounds. Standard generative models often create novel molecules that require months to synthesize or prove synthesizable altogether. Calici avoids these delays by focusing on stock chemical libraries from established vendor partners. If the AI recommends a compound, it is already available, synthesizable, and can be shipped to the client’s lab immediately.

Behind this is a big data search AI strategy using reinforcement learning agents. Instead of scanning hundreds of millions of compounds one by one, these agents analyze the chemical space, identify patterns, and focus searches where the best candidates are most likely to be found. Pharmaco-Net^® draws from an extensive virtual compound library of over 600 million molecules, sourced through partnerships with TargetMol, Mcule, and Enamine, enabling rapid in-silico screening and real-world feasibility.

From Digital Prediction to Laboratory Reality

Pharmaco-Net^®, a drug development platform developed by Calici, has supported a wide range of projects, from academic research to industry drug development. One example comes from Pohang University, where a team was investigating castration-resistant prostate cancer, a form resistant to existing treatments in up to 15 percent of cases. Using Pharmaco-Net^®, the researchers screened synthesizable libraries against their target protein, ETV5, and identified Obeticholic acid, an FDA-approved drug for another indication. In mouse models, the compound significantly reduced tumor size, leading to a PNAS-published study.

While AI is a powerful enabler, Calici views drug discovery as a team sport. The company works closely with institutions such as Korea University and Yeungnam University, combining technical tools with human expertise. As the founder explains, advanced AI is like a high-end golf club. It can improve your game, but success depends on the skill of the player and the support of the team around them.

By making AI accessible, actionable, and focused on real-world results, Pharmaco-Net^® is bridging the gap between computational predictions and laboratory breakthroughs. Its emphasis on synthesizable compounds, user-friendly tools, and collaborative science is helping researchers move faster from idea to therapy and ultimately bringing better medicines to patients sooner.