Novel Reagent Space: Identifying Unorderable but Readily Synthesizable Building Blocks

Artificial Intelligence Methods and Models for Retro-Biosynthesis: A Scoping Review

Retrosynthesis aims to efficiently plan the synthesis of desirable chemicals by strategically breaking down molecules into readily available building block compounds. Having a long history in chemistry, retro-biosynthesis has also been used in the fields of biocatalysis and synthetic biology. Artificial intelligence (AI) is driving us toward new frontiers in synthesis planning and the exploration of chemical spaces, arriving at an opportune moment for promoting bioproduction that would better align with green chemistry, enhancing environmental practices. In this review, we summarize the recent advancements in the application of AI methods and models for retrosynthetic and retro-biosynthetic pathway design. These techniques can be based either on reaction templates or generative models and require scoring functions and planning strategies to navigate through the retrosynthetic graph of possibilities. We finally discuss limitations and promising research directions in this field.

Unveiling the Quantitative Relationships between Electron Distribution and Steric Hindrance of Organic Amines and Their Reaction Rates with Carbonyl Sulfur: A Theoretical Calculation Investigation

The removal of carbonyl sulfide (COS) commonly contained in natural gas is of great significance but still very challenging via a widely employed absorption process due to its low reactivity and solubility in various commercial solvents. Artificial intelligence (AI) is playing an increasingly important role in the exploration of desulfurization solvents. However, practically feasible AI models still lack a thorough understanding of the reaction mechanisms. Machine learning (ML) models established on chemical mechanisms exhibit enhanced chemical interpretability and prediction performance. In this study, we constructed a series of solvent molecules with varying functional groups, including linear aliphatic amines, cyclic aliphatic amines, and aromatic amines and proposed a three-step reaction pathway to dissect the effects of charge and steric hindrance of different substituents on their reaction rates with COS. Chemical descriptors, based on electrostatic potential (ESP), average local ionization energy (ALIE) theory, Hirshfeld charges, and Fukui functions, were used to correlate and predict the electrophilic reactivity of amine groups with COS. Substituents influence the reaction rate by changing the attraction interaction of amine groups to COS molecules and the electron rearrangement in the electrophilic reaction. Furthermore, they have more pronounced steric effects on the reaction rate in the linear amines. The descriptors N_ALIE and q(N) were found to be crucial in predicting the reactivity of amine groups with COS. Present study provides a comprehensive understanding of the reaction mechanisms of COS with amine compounds, offers specific chemical principles for the development of chemistry-driven ML models, sheds light on other types of electrophilic reactions occurring on amine and phosphine groups, and guides the development of chemical solvents in gas absorption processes.

Enabling Deoxygenative C(sp2)-C(sp3) Cross-Coupling for Parallel Medicinal Chemistry

Herein we report the development of an automated deoxygenative C(sp2)-C(sp3) coupling of aryl bromide with alcohols to enable parallel medicinal chemistry. Alcohols are among the most diverse and abundant building blocks, but their usage as alkyl precursors has been limited. Although metallaphotoredox deoxygenative coupling is becoming a promising strategy to form C(sp2)-C(sp3) bond, the reaction setup limits its widespread application in library synthesis. To achieve high throughput and consistency, an automated workflow involving solid-dosing and liquid-handling robots has been developed. We have successfully demonstrated this high-throughput protocol is robust and consistent across three automation platforms. Furthermore, guided by cheminformatic analysis, we examined alcohols with comprehensive chemical space coverage and established a meaningful scope for medicinal chemistry applications. By accessing the rich diversity of alcohols, this automated protocol has the potential to substantially increase the impact of C(sp2)-C(sp3) cross-coupling in drug discovery.