DeepReac+: deep active learning for quantitative modeling of organic chemical reactions

Abstract

Various computational methods have been developed for quantitative modeling of organic chemical reactions, however, the lack of universality as well as the requirement of large amounts of experimental data have...

Uncertainty quantification: Can we trust artificial intelligence in drug discovery?

“…Owing to the time- and resource-intensive nature of biological and chemical experiments, how to generate new data to improve model performance more efficiently is a key problem in drug discovery ( Yu et al., 2021 ). To address this issue, active learning (AL), an uncertainty-guided algorithm, has begun to show promise and has increasingly been used ( Ding et al., 2021 ; Gong et al., 2021 ; Jansen et al., 2019 ; Yang et al., 2021 ). In AL, a model is typically initialized with a limited training set (e.g., currently available samples).…”

AROPS: A Framework of Automated Reaction Optimization with Parallelized Scheduling

“…Benchmark C (Scheme 1C) is a Suzuki-Miyaura C-C coupling reaction dataset reported by Pfizer, 5 which was adopted in many works because of its data integrity. 31,41,42 The design space…”

Navigating chemical reaction space – application to DNA-encoded chemistry

“…1B). MEGAN (Molecule Edit Graph Attention Network) from the Jastrze ˛bski group 25 and DeepReact + by Gong et al 26 stand out for reaction conditions optimization, while the landmark research works of Segler et al 27 and Chematica from the Grzybowski group, as well as MEGAN, support retro-synthetic analysis. 25,28 32 Lastly, in the area of library design, two tools shall be mentioned: eDESIGNER, developed in 2020 by Martìn et al, 33 and Synthl, presented by Zabolotna et al 34 The former uses established reactions in the DEL eld, and the latter, though being useful for building block-based library design, is not tailored to DEL chemistry.…”