The platform combines artificial intelligence and molecular modeling tools to enable early-stage development of drug candidates — from design hypotheses to the selection of molecules for synthesis and biological testing.

The Russian startup Ligand Pro, founded by professors and a PhD student from Skoltech (part of the VEB.RF group), has introduced a platform for drug molecule development that uses artificial intelligence and molecular modeling methods. The solution brings together molecular design, computational experiments, and analysis of their results in a single digital environment.

The platform is designed for the computational stage of early drug development. With its help, researchers can formulate and test molecular design hypotheses, generate and optimize structures, predict their interaction with the target protein, assess physicochemical properties and structural risks, and analyze the patent landscape and synthetic accessibility of compounds. After a series of calculations, the user can create a shortlist of the most promising molecules for subsequent synthesis and experimental testing.

“This is the first in Russia platform that combines AI and molecular modeling tools in a single environment for the full computational cycle of early drug development. With the platform, we are making advanced AI tools accessible to pharmaceutical companies and research teams so they can accelerate the development of new drug molecules, improve the efficiency of computational experiments, and move faster from hypotheses to promising candidate molecules,” said Marina Pak, co-founder and CEO of Ligand Pro, a graduate of Skoltech’s MSc and PhD programs.

Unlike standalone computational services, Ligand Pro is organized as a working environment for drug development projects. In it, molecules, protein targets, compound collections, design hypotheses, and calculation results are interconnected. This allows researchers not only to run individual models but also to preserve the logic of decision-making across multiple cycles of molecular design.

“Today many teams use separate programs for different stages of development: one tool for docking, another for property prediction, a third for generating or analyzing molecules. The results have to be manually combined and interpreted, and the context of why a particular structure was created often remains outside the computational system. We wanted to build not a catalog of online tools but a unified molecular design environment. In Ligand Pro, a researcher can formulate a hypothesis, link a series of molecules to it, test it through calculations, and then select candidates based on the combined data,” said Sergey Nikolenko, head of chemoinformatics at Ligand Pro.

The platform offers modern AI and molecular modeling tools for early-stage drug development tasks. These include generative models for creating and optimizing new chemical structures, retrosynthetic analysis tools, calculation of molecular descriptors and structural alerts, and the flagship Matcha program — a highly accurate and fast AI-based molecular docking tool that predicts the position of a small molecule in a complex with a target protein.

The platform is intended primarily for specialists in medicinal and computational chemistry, as well as research teams at pharmaceutical companies and academic organizations working on the discovery and optimization of bioactive compounds.

Ligand Pro is currently in closed beta testing. Representatives of pharmaceutical companies and academic teams involved in drug molecule design are joining the platform. At this stage, the team is gathering feedback on use cases and the interface and continues to add new computational modules. After testing is complete, access to the platform is planned to be opened to a wider range of users.

The platform is available at the Ligand Pro website.