SynEPD: a curated mechanistic data resource with electron-pushing annotations and hierarchical taxonomy for polar organic reactions
The paper introduces SynEPD, a curated dataset of 1,926 mass- and charge-balanced polar organic reactions that uniquely integrates hierarchical taxonomy, atom-mapped transformations, and structured electron-pushing annotations to enable mechanistic retrieval, template mining, and the development of predictive models for electron flow.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Chemistry is often taught as a study of static structures, but at its heart, it is a story of movement. It is the story of how atoms rearrange themselves to form new substances, a process driven by the flow of electrons. In the classroom, scientists and students have long used a specific visual language to track this invisible traffic: curved arrows. These arrows, drawn on paper, show how pairs of electrons move from one location to another, breaking old bonds and forming new ones. This method, known as electron pushing, is the standard way to explain why a reaction happens and how it proceeds step by step. However, while these diagrams are the common language for teaching and reasoning about chemical reactions, they have rarely been stored in a way that computers can easily read, check, or learn from. Most large databases of chemical reactions record only the starting materials and the final products, leaving the crucial journey—the path the electrons took—unwritten. Without this detailed map of the journey, it is difficult to build computer programs that can truly understand how reactions work or predict new ones with high accuracy.
A team of researchers has now created a solution to this gap by building a new, carefully curated collection of chemical reactions that includes these detailed electron-flow maps. They call this resource SynEPD. Instead of just listing what goes in and what comes out, this database records every single step of the electron movement for nearly two thousand different polar organic reactions. These are reactions where electrons shift between atoms to create charged intermediates, a category that covers a vast amount of the chemistry used in making medicines, plastics, and other materials. The researchers did not simply scrape this data from existing sources; they manually constructed and verified each entry to ensure it was chemically perfect. They checked that the number of atoms and the electrical charge were balanced on both sides of the reaction, and they confirmed that the sequence of electron movements they recorded actually transformed the starting molecules into the correct final products.
The result is a library containing 1,926 distinct reactions, each linked to a clear, hierarchical system that sorts them by how they work. For every reaction in the set, the team recorded 8,123 individual electron-pushing arrows. These arrows are not just pictures; they are encoded instructions that tell a computer exactly which atom loses an electron pair and which atom gains it. The database also includes a unique way of looking at the "heart" of each reaction, identifying the specific atoms that change their connections and the temporary states they pass through. This allows researchers to group reactions not just by what they look like, but by the specific logic of their electron flow. To make this data useful for the wider scientific community, the team also created a bridge connecting their new system to an existing standard for named chemical reactions, allowing scientists to find these detailed maps using familiar names.
The creation of this resource involved a rigorous process of human review and automated checking. Experts in organic chemistry selected representative examples for different types of reactions, drawing the electron-flow diagrams by hand and then translating them into a digital format. These diagrams were then subjected to a strict test: a computer program replayed every single arrow in the sequence to see if it successfully turned the starting molecule into the final product. If the sequence failed to produce the correct result, the entry was rejected. This ensured that every record in the database is a valid, working model of a chemical transformation. The team also organized the reactions into a structured tree, starting with broad families like acid-base chemistry or substitution, and drilling down into specific, named transformations. This hierarchy helps researchers find exactly the kind of mechanism they are looking for, whether they are studying a simple proton transfer or a complex multi-step rearrangement.
By separating the overall change in the molecule from the specific order of electron movements, SynEPD offers a new way to study chemical reactivity. It allows scientists to ask questions that were previously difficult to answer, such as how many different ways electrons can move to achieve the same result, or which specific patterns of electron flow are most common in nature. The database is now available for anyone to use, providing a foundation for developing smarter computer models that can predict how molecules will react. It represents a shift from simply cataloging chemical outcomes to understanding the dynamic rules that govern them. With this resource, the invisible dance of electrons becomes a visible, searchable, and testable record, offering a clearer view of the fundamental processes that drive chemical change.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.