Recently, many research groups have been addressing data-driven approaches for (retro)synthetic reaction prediction and retrosynthetic analysis. Although the performances of the data-driven approach have progressed because of recent advances of machine learning and deep learning techniques, problems such as improving capability of reaction prediction and the black-box problem of neural networks persist for practical use by chemists. To spread data-driven approaches to chemists, we focused on two challenges: improvement of retrosynthetic reaction prediction and interpretability of the prediction. In this paper, we propose an interpretable prediction framework using graph convolutional networks (GCN) for retrosynthetic reaction prediction and integrated gradients (IG) for visualization of contributions to the prediction to address these challenges. As a result, from the viewpoint of balanced accuracies, our model showed better performances than the approach using an extended-connectivity fingerprint. Furthermore, IG-based visualization of the GCN prediction successfully highlighted reaction-related atoms.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1021/acs.jcim.9b00538 | DOI Listing |
Infect Drug Resist
December 2024
Department of Cardiovascular Medicine, Showa University Fujigaoka Hospital, Yokohama, Japan.
Background: The emergence of the Omicron variant of severe acute respiratory syndrome coronavirus-2 has significantly altered the clinical features and severity of coronavirus disease 2019 (COVID-19).
Objective: This study aims to evaluate whether the clinical factors that previously predicted COVID-19 remain valid following the emergence of the Omicron variant.
Methods: This cross-sectional study was conducted at Showa University Fujigaoka Hospital from April 2022 to March 2023.
J Phys Chem B
December 2024
Collaborative Innovation Center of Chemical Science and Engineering, Tianjin University, Tianjin 300072, China.
As a predictive tool, quantum chemical calculations can be used to design protic ionic liquids (PILs) and predict the result. By adding anionic negative potential sites, two dual-functional PILs diethylenetriamine-barbituric acid [CHN][CHNO] and diethylenetriamine-ethylenolactonium [CHN][CHNO] were designed. The simulation results indicated that multisite absorption of anions and cations resulted in an expected absorption ratio exceeding 3:1 (mol CO:mol ILs).
View Article and Find Full Text PDFNanoscale
December 2024
Institute of Chemistry, Inorganic Materials Laboratory 52S, The Islamia University of Bahawalpur-63100, Pakistan.
As a future fuel, obtaining hydrogen from water could be a game changer for the renewable energy sector, because it has the potential to be used as an alternative to fossil fuels. The current project has been designed to develop catalysts that can produce hydrogen from water on irradiation by sunlight. For this purpose, CdS, Cu/CdS, Pd/CdS, and Cu-Pd/CdS catalysts were successfully synthesised and utilized for hydrogen generation.
View Article and Find Full Text PDFSmall
December 2024
Institute of Light and Matter, University Claude Bernard Lyon 1, CNRS, UMR5306, Villeurbanne, F-69622, France.
Bimetallic nanoparticles are used in numerous applications in catalysis, plasmonics or fuel cell technology. The addition of the second metal to the nanoparticles allows enhancing and fine-tuning their properties by choosing their composition, size, shape and environment. However, the crucial additional parameter of chemical structure within the particle is difficult to predict and access experimentally, even though segregated core-shell structures and random alloys can have drastically different physicochemical properties.
View Article and Find Full Text PDFTransfusion
December 2024
School of Biomedical Sciences, Faculty of Health, University of Plymouth, Plymouth, UK.
Background: The Rh blood group system (ISBT004) is encoded by two homologous genes, RHD and RHCE. Polymorphism in these two genes gives rise to 56 antigens, which are highly immunogenic and clinically significant. This study extended previous work on the establishment of RHD allele specific reference sequences using next generation sequencing (NGS) with the Ion Personal Genome Machine (Ion PGM) to sequence the complete RHCE gene.
View Article and Find Full Text PDFEnter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!