A universal system for digitization and automatic execution of the chemical synthesis literature

Abstract

Robotic systems for chemical synthesis are growing in popularity but can be difficult to run and maintain because of the lack of a standard operating system or capacity for direct access to the literature through natural language processing. Here we show an extendable chemical execution architecture that can be populated by automatically reading the literature, leading to a universal autonomous workflow. The robotic synthesis code can be corrected in natural language without any programming knowledge and, because of the standard, is hardware independent. This chemical code can then be combined with a graph describing the hardware modules and compiled into platform-specific, low-level robotic instructions for execution. We showcase automated syntheses of 12 compounds from the literature, including the analgesic lidocaine, the Dess-Martin periodinane oxidation reagent, and the fluorinating agent AlkylFluor.

Navigating through the Maze of Homogeneous Catalyst Design with Machine Learning

“…Moreover, central are efforts to digitize published reaction protocols with translation algorithms such as SynthReader and to standardize the description and execution of syntheses on automated systems with the Chemical Description Language XDL [69]. We envision that comprehensive digitization of chemistry will lead to protocols being deposited in standardized databases, similar to how crystal structures are deposited in the CSD, allowing the reproduction of experiments by downloading files and executing them on synthesis machines.…”

“…While considerable successes toward the automation of organic retrosynthesis and forward synthesis prediction [70] have been achieved, similar efforts to predict synthesis procedures for new molecules are largely unheard of. We envision that combining forward synthesis prediction with ML language models [71] and standardized synthesis descriptions like XDL [69] could provide an entry toward that goal. While these are glimpses of how this could be achieved in organic chemistry, to the best of our knowledge, for organometallic or inorganic chemistry, the necessary tools for retrosynthesis and forward synthesis prediction do not exist.…”

Natural Products, the Fourth Industrial Revolution, and the Quintuple Helix

“…52 A robotic system which reads a synthetic method from the literature and automatically translates that into a reaction has been described. 53 In addition, a combined expert (with reaction rules) and machine learning (literature data) approach showed that the processes can be synergistic in improving overall synthetic accuracy. 54 Collaboration between machine learning scientists and bench chemists will be important for further advances in predictable, high-yield pathway outcomes.…”

Deep learning and generative methods in cheminformatics and chemical biology: navigating small molecule space intelligently

“…This clearly applies to problems in chemistry and chemical biology that involve navigating large search spaces of molecules, and intelligent automation has been an important subset of activities here (e.g. [172][173][174][175][176][177][178][179][180][181][182]). 'Active learning' describes the kinds of methods that use knowledge of existing data to determine where best to explore next, and is normally used to balance exploration (looking for promising regions of the search space) with exploitation (a promising local search) [183].…”