The Automatic Detection of Chemical Reaction Sites

Benchmarking machine-readable vectors of chemical reactions on computed activation barriers

In recent years, there has been a surge of interest in predicting computed activation barriers, to enable the acceleration of the automated exploration of reaction networks. Consequently, various predictive approaches have emerged, ranging from graph-based models to methods based on the three-dimensional structure of reactants and products. In tandem, many representations have been developed to predict experimental targets, which may hold promise for barrier prediction as well. Here, we bring together all of these efforts and benchmark various methods (Morgan fingerprints, the DRFP, the CGR representation-based Chemprop, SLATMd, B2Rl2, EquiReact and language model BERT + RXNFP) for the prediction of computed activation barriers on three diverse datasets.

Learning symmetry-aware atom mapping in chemical reactions through deep graph matching

Accurate atom mapping, which establishes correspondences between atoms in reactants and products, is a crucial step in analyzing chemical reactions. In this paper, we present a novel end-to-end approach that formulates the atom mapping problem as a deep graph matching task. Our proposed model, AMNet (Atom Matching Network), utilizes molecular graph representations and employs various atom and bond features using graph neural networks to capture the intricate structural characteristics of molecules, ensuring precise atom correspondence predictions. Notably, AMNet incorporates the consideration of molecule symmetry, enhancing accuracy while simultaneously reducing computational complexity. The integration of the Weisfeiler-Lehman isomorphism test for symmetry identification refines the model's predictions. Furthermore, our model maps the entire atom set in a chemical reaction, offering a comprehensive approach beyond focusing solely on the main molecules in reactions. We evaluated AMNet's performance on a subset of USPTO reaction datasets, addressing various tasks, including assessing the impact of molecular symmetry identification, understanding the influence of feature selection on AMNet performance, and comparing its performance with the state-of-the-art method. The result reveals an average accuracy of 97.3% on mapped atoms, with 99.7% of reactions correctly mapped when the correct mapped atom is within the top 10 predicted atoms.Scientific contributionThe paper introduces a novel end-to-end deep graph matching model for atom mapping, utilizing molecular graph representations to capture structural characteristics effectively. It enhances accuracy by integrating symmetry detection through the Weisfeiler-Lehman test, reducing the number of possible mappings and improving efficiency. Unlike previous methods, it maps the entire reaction, not just main components, providing a comprehensive view. Additionally, by integrating efficient graph matching techniques, it reduces computational complexity, making atom mapping more feasible.

Precise atom-to-atom mapping for organic reactions via human-in-the-loop machine learning

Atom-to-atom mapping (AAM) is a task of identifying the position of each atom in the molecules before and after a chemical reaction, which is important for understanding the reaction mechanism. As more machine learning (ML) models were developed for retrosynthesis and reaction outcome prediction recently, the quality of these models is highly dependent on the quality of the AAM in reaction datasets. Although there are algorithms using graph theory or unsupervised learning to label the AAM for reaction datasets, existing methods map the atoms based on substructure alignments instead of chemistry knowledge. Here, we present LocalMapper, an ML model that learns correct AAM from chemist-labeled reactions via human-in-the-loop machine learning. We show that LocalMapper can predict the AAM for 50 K reactions with 98.5% calibrated accuracy by learning from only 2% of the human-labeled reactions from the entire dataset. More importantly, the confident predictions given by LocalMapper, which cover 97% of 50 K reactions, show 100% accuracy for 3,000 randomly sampled reactions. In an out-of-distribution experiment, LocalMapper shows favorable performance over other existing methods. We expect LocalMapper can be used to generate more precise reaction AAM and improve the quality of future ML-based reaction prediction models.

Extraction of organic chemistry grammar from unsupervised learning of chemical reactions

Humans use different domain languages to represent, explore, and communicate scientific concepts. During the last few hundred years, chemists compiled the language of chemical synthesis inferring a series of "reaction rules" from knowing how atoms rearrange during a chemical transformation, a process called atom-mapping. Atom-mapping is a laborious experimental task and, when tackled with computational methods, requires continuous annotation of chemical reactions and the extension of logically consistent directives. Here, we demonstrate that Transformer Neural Networks learn atom-mapping information between products and reactants without supervision or human labeling. Using the Transformer attention weights, we build a chemically agnostic, attention-guided reaction mapper and extract coherent chemical grammar from unannotated sets of reactions. Our method shows remarkable performance in terms of accuracy and speed, even for strongly imbalanced and chemically complex reactions with nontrivial atom-mapping. It provides the missing link between data-driven and rule-based approaches for numerous chemical reaction tasks.

Automatic mapping of atoms across both simple and complex chemical reactions

Mapping atoms across chemical reactions is important for substructure searches, automatic extraction of reaction rules, identification of metabolic pathways, and more. Unfortunately, the existing mapping algorithms can deal adequately only with relatively simple reactions but not those in which expert chemists would benefit from computer's help. Here we report how a combination of algorithmics and expert chemical knowledge significantly improves the performance of atom mapping, allowing the machine to deal with even the most mechanistically complex chemical and biochemical transformations. The key feature of our approach is the use of few but judiciously chosen reaction templates that are used to generate plausible "intermediate" atom assignments which then guide a graph-theoretical algorithm towards the chemically correct isomorphic mappings. The algorithm performs significantly better than the available state-of-the-art reaction mappers, suggesting its uses in database curation, mechanism assignments, and - above all - machine extraction of reaction rules underlying modern synthesis-planning programs.

Information Retrieval and Text Mining Technologies for Chemistry

Efficient access to chemical information contained in scientific literature, patents, technical reports, or the web is a pressing need shared by researchers and patent attorneys from different chemical disciplines. Retrieval of important chemical information in most cases starts with finding relevant documents for a particular chemical compound or family. Targeted retrieval of chemical documents is closely connected to the automatic recognition of chemical entities in the text, which commonly involves the extraction of the entire list of chemicals mentioned in a document, including any associated information. In this Review, we provide a comprehensive and in-depth description of fundamental concepts, technical implementations, and current technologies for meeting these information demands. A strong focus is placed on community challenges addressing systems performance, more particularly CHEMDNER and CHEMDNER patents tasks of BioCreative IV and V, respectively. Considering the growing interest in the construction of automatically annotated chemical knowledge bases that integrate chemical information and biological data, cheminformatics approaches for mapping the extracted chemical names into chemical structures and their subsequent annotation together with text mining applications for linking chemistry with biological information are also presented. Finally, future trends and current challenges are highlighted as a roadmap proposal for research in this emerging field.

Information retrieval research in the Department of Information Studies, University of Sheffield: 1965-1985

This paper discusses research which was carried out at the Department of Information Studies, University of Sheffield in the period 1965 to 1985 into storage and retrieval techniques for databases of textual and chemical structure data. The research includes the development of methods for the auto matic production of printed subject indexes and for the inde xing and retrieval of chemical structures and chemical reac tions, the variety generation method for the analysis, characterization and storage of data in a range of types of textual database, the prediction of biological activity in chemical compounds, and the design of document retrieval systems.

Atom mapping with constraint programming

Chemical reactions are rearrangements of chemical bonds. Each atom in an educt molecule thus appears again in a specific position of one of the reaction products. This bijection between educt and product atoms is not reported by chemical reaction databases, however, so that the “Atom Mapping Problem” of finding this bijection is left as an important computational task for many practical applications in computational chemistry and systems biology. Elementary chemical reactions feature a cyclic imaginary transition state (ITS) that imposes additional restrictions on the bijection between educt and product atoms that are not taken into account by previous approaches. We demonstrate that Constraint Programming is well-suited to solving the Atom Mapping Problem in this setting. The performance of our approach is evaluated for a manually curated subset of chemical reactions from the KEGG database featuring various ITS cycle layouts and reaction mechanisms.

Automated reaction mapping

Automated reaction mapping is a fundamental first step in the analysis of chemical reactions and opens the door to the development of sophisticated chemical kinetic tools. This article formulates the reaction mapping problem as an optimization problem. The problem is shown to be NP-Complete for general graphs. Five algorithms based on canonical graph naming and enumerative combinatoric techniques are developed to solve the problem. Unlike previous formulations based on limited configurations or classifications, our algorithms are uniquely capable of mapping any reaction that can be represented as a set of chemical graphs optimally. This is due to the direct use of Graph Isomorphism as the basis for these algorithms as opposed to the more commonly used Maximum Common Subgraph. Experimental results on chemical and biological reaction databases demonstrate the efficiency of our algorithms.

From chemical documentation to chemoinformatics: 50 years of chemical information science

This paper summarizes the historical development of the discipline that is now called `chemoinformatics'. It shows how this has evolved, principally as a result of technological developments in chemistry and biology during the past decade, from long-established techniques for the modelling and searching of chemical molecules. A total of 30 papers, the earliest dating back to 1957, are briefly summarized to highlight some of the key publications and to show the development of the discipline.

Construction of a generic reaction knowledge base by reaction data mining

As synthesis by combinatorial chemistry and high throughput screening have become well-established strategies in the drug discovery process, chemists face increased challenges in managing large amounts of data and using these data to design more diverse and focused libraries. As synthesis is an intuitive and empirical process, however, the classical approaches to computer-assisted synthesis planning do not fully satisfy the needs of the synthetic chemist. We describe a novel computational technique for extracting reaction data and building a generic reaction knowledge base (GRKB) to provide chemists with useful and well-organized knowledge. The method consists of three key steps: (1) the automatic recognition of reaction centers, (2) the definition of a hierarchy of reaction patterns, and (3) the organization of the generic reaction knowledge. Significant reaction knowledge has been discovered via mining a subset of the InfoChem Reaction database. A frame system has been constructed to store and retrieve the GRKB. Applications of this GRKB to synthesis planning are illustrated.