New insight into atomic-level interpretation of interactions in molecules and reacting systems.

Phys Chem Chem Phys

Institut de Chimie Moléculaire de Reims UMR CNRS 7312, Université de Reims Champagne-Ardenne, Moulin de la Housse, F-51687 Reims Cedex 02 BP 1039, France.

Published: April 2023

We hereby introduce the atomic degree of interaction (DOI), a new concept rooted in the electron density-based independent gradient model (IGM). Capturing any manifestation of electron density sharing around an atom, including covalent and non-covalent situations, this index reflects the attachment strength of an atom to its molecular neighbourhood. It is shown to be very sensitive to the local chemical environment of the atom. No significant correlation could be found between the atomic DOI and various other atomic properties, making this index a specific source of information. However, examining the simple H + H reacting system, a strong connection has been established between this electron density-based index and the scalar reaction path curvature, the cornerstone of the benchmark unified reaction valley approach (URVA). We observe that reaction path curvature peaks appear when atoms experience an acceleration phase of electron density sharing during the reaction, detected by peaks of the DOI second derivative either in the forward or reverse direction. This new IGM-DOI tool is only in its early stages, but it opens the way to an atomic-level interpretation of reaction phases. More generally, the IGM-DOI tool may also serve as an atomic probe of electronic structure changes of a molecule under the influence of physicochemical perturbations.

Download full-text PDF

Source
http://dx.doi.org/10.1039/d2cp02839eDOI Listing

Publication Analysis

Top Keywords

atomic-level interpretation
8
electron density-based
8
electron density
8
density sharing
8
reaction path
8
path curvature
8
igm-doi tool
8
reaction
5
insight atomic-level
4
interpretation interactions
4

Similar Publications

Atomic-level simulations are widely used to study biomolecules and their dynamics. A common goal in such studies is to compare simulations of a molecular system under several conditions-for example, with various mutations or bound ligands-in order to identify differences between the molecular conformations adopted under these conditions. However, the large amount of data produced by simulations of ever larger and more complex systems often renders it difficult to identify the structural features that are relevant to a particular biochemical phenomenon.

View Article and Find Full Text PDF

Machine Learning for Reaction Performance Prediction in Allylic Substitution Enhanced by Automatic Extraction of a Substrate-Aware Descriptor.

J Chem Inf Model

January 2025

Department of Computer Science and Engineering, and Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China.

Despite remarkable advancements in the organic synthesis field facilitated by the use of machine learning (ML) techniques, the prediction of reaction outcomes, including yield estimation, catalyst optimization, and mechanism identification, continues to pose a significant challenge. This challenge arises primarily from the lack of appropriate descriptors capable of retaining crucial molecular information for accurate prediction while also ensuring computational efficiency. This study presents a successful application of ML for predicting the performance of Ir-catalyzed allylic substitution reactions.

View Article and Find Full Text PDF

Time-resolved serial crystallography at X-ray Free Electron Lasers offers the opportunity to observe ultrafast photochemical reactions at the atomic level. The technique has yielded exciting molecular insights into various biological processes including light sensing and photochemical energy conversion. However, to achieve sufficient levels of activation within an optically dense crystal, high laser power densities are often used, which has led to an ongoing debate to which extent photodamage may compromise interpretation of the results.

View Article and Find Full Text PDF

Accurately identifying sites of metabolism (SoM) mediated by cytochrome P450 (CYP) enzymes, which are responsible for drug metabolism in the body, is critical in the early stage of drug discovery and development. Current computational methods for CYP-mediated SoM prediction face several challenges, including limitations to traditional machine learning models at the atomic level, heavy reliance on complex feature engineering, and the lack of interpretability relevant to medicinal chemistry. Here, we propose GraphCySoM, a novel molecule-level modeling approach based on graph neural networks, utilizing lightweight features and interpretable annotations on substructures, to effectively and interpretably predict CYP-mediated SoM.

View Article and Find Full Text PDF

DeepRSMA: a cross-fusion-based deep learning method for RNA-small molecule binding affinity prediction.

Bioinformatics

November 2024

Machine Intellection Department, Institute for Infocomm Research, Agency for Science, Technology and Research (A*STAR), Singapore 138632, Singapore.

Article Synopsis
  • - This study introduces DeepRSMA, a new deep learning method designed to predict RNA-small molecule affinity (RSMA), which is vital for developing RNA-targeted drugs for diseases.
  • - DeepRSMA utilizes advanced techniques like nucleotide-level and atomic-level feature extraction, as well as a Transformer-based cross-fusion module, to effectively analyze RNA and small molecules from different perspectives.
  • - The results indicate that DeepRSMA outperforms existing methods, providing insights that could assist in designing effective RNA-targeted therapies, with the code and data accessible for public use.
View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!