System and method for feedback-driven automated drug discovery
Inventors
Norvaisas, Povilas • Tal, Roy • Jocys, Zygimantas • Knuff, Charles Dazler • Prat, Alvaro • Kamuntavicius, Gintautas • Aty, Hisham Abdel • Bastas, Orestis • Nonkovic, Nikola
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12248885-B2
Publication Date
2025-03-11
Expiration Date
Abstract
A system and method for feedback-driven automated drug discovery which combines machine learning algorithms with automated research facilities and equipment to make the process of drug discovery more data driven and less reliant on intuitive decision-making by experts. In an embodiment, the system comprises automated research equipment configured to perform automated assays of chemical compounds, a data platform comprising drug databases and an analysis engine, a bioactivity and de novo modules operating on the data platform, and a retrosynthesis system operating on the drug discovery platform, all configured in a feedback loop that drives drug discovery by using the outcome of assays performed on the automated research equipment to feed the bioactivity module and retrosynthesis systems, which identify new molecules for testing by the automated research equipment.
Core Innovation
The invention provides a feedback-driven automated drug discovery system that combines an iterative physical testing system with computing and artificial intelligence assistance. Automated biochemical assay equipment receives an amount of a drug candidate molecule and an assay procedure including a first parameter, performs an assay to produce an assay result indicating whether the first parameter has been met, and sends the assay result to a laboratory API together with chemical notation for the drug candidate molecule.
A bioactivity module receives assay procedures and generates a set of possible drug candidates, and sends chemical notation for the set of possible drug candidates to the automated biochemical assay equipment for performance of primary assays. In parallel, a de novo ligand discovery module receives the primary assay result, the drug candidate molecule, and a binding site from a chemical database, and performs one or more interpolations or perturbations of the drug candidate molecule to generate a new drug candidate molecule predicted to meet or exceed the first parameter.
The de novo ligand discovery module sends chemical notation for the new drug candidate molecule to a retrosynthesis system, which performs retrosynthesis using rules for retrosynthesis of molecules stored in the chemical database and produces chemical notation for precursors and synthesis steps for synthesizing the new drug candidate molecule. Laboratory equipment orchestration software, including a laboratory API module, a laboratory workflow control module, and a laboratory equipment utilization planning module, coordinates receipt of instructions and experimental workflows, including primary assays, secondary assay, and outputting a result.
Claims Coverage
The provided independent claims are clm-00001 and clm-00007. Across these claims, three inventive features are presented: automated biochemical assays linked to AI-assisted candidate generation, retrosynthesis from chemical database rules, and laboratory equipment orchestration through a laboratory API.
Feedback-driven automated drug discovery system
An iterative physical testing system with automated biochemical assay equipment that receives a drug candidate molecule and assay procedure comprising a first parameter, performs an assay to produce an assay result indicating whether the first parameter has been met, and sends the assay result to a bioactivity module via a laboratory API along with chemical notation. A chemical synthesizer receives synthesis instructions from a laboratory equipment controller, synthesizes a new drug candidate molecule from a supply of chemicals using chemical notation of precursors and synthesis steps provided by a retrosynthesis system, and provides an amount of the new drug candidate molecule to the automated biochemical assay equipment.
Feedback-driven automated drug discovery method
Using automated biochemical assay equipment to receive an amount of a drug candidate molecule and an assay procedure comprising a first parameter, automatically perform an assay to produce an assay result indicating whether the first parameter has been met, and send the assay result to a bioactivity module along with chemical notation and a binding site when met; using a chemical synthesizer to synthesize a new drug candidate molecule from a supply of chemicals using chemical notation of precursors and synthesis steps provided by a retrosynthesis system; using laboratory equipment orchestration software to control the system via a laboratory API module and a laboratory workflow control module defining experimental workflows; using a bioactivity module to generate a set of possible drug candidates; using a de novo ligand discovery module to receive a primary assay result, drug candidate molecule, and binding site, and perform one or more interpolations or perturbations to generate a new drug candidate molecule predicted to meet or exceed the first parameter; and using a retrosynthesis system to perform retrosynthesis using rules for retrosynthesis of molecules stored in a chemical database.
The independent claims define a feedback-driven automated drug discovery system and method that iteratively link automated biochemical assays to a bioactivity module and a de novo ligand discovery module, and to a retrosynthesis system that provides chemical notation for precursors and synthesis steps based on rules in a chemical database, with laboratory equipment orchestration software controlling workflows through a laboratory API.
Stated Advantages
Documented Applications
No documented applications found
Interested in licensing this patent?