(95c) AI-Driven Hypergraph Network of Organic Chemistry: Applications in Reaction Classification

Advancements in high throughput screening, accessibility to complex chemical design space, and accurate molecular modeling methods have led to an accelerated discovery of new reactions and molecules in recent years. To understand recent trends and identify possible future trajectories, a holistic study of these reactions is required. Studying the network of organic chemistry (a huge network containing all known reactions) as a graph where molecules are represented as nodes and reactions as edges allows for the application of methods from graph theory to study them. As a result, several network theory-based studies have been reported that use a directed graph representation of chemical reactions. In this work, we perform a study based on representing chemical reactions as hypergraphs where the nodes represent the participating molecules and hyperedges represent reactions between nodes [1]. We use a standard reaction dataset to construct a hypergraph network of organic chemistry and report its statistics such as degree distribution, average path length, assortativity or degree correlations, PageRank centrality, and graph-based clusters (or communities). We also compute each statistic for an equivalent directed graph representation of reactions to draw parallels and highlight differences between the two. To demonstrate the AI applicability of hypergraph reaction representation, we generate dense hypergraph embeddings and use them in the reaction classification problem. We conclude that the hypergraph representation is flexible, preserves reaction context, and uncovers hidden insights that are otherwise not apparent in a traditional directed graph representation of chemical reactions [1-4].

References:

1. Mann, Vipul, and Venkat Venkatasubramanian. "AI-driven hypergraph network of organic chemistry: network statistics and applications in reaction classification." Reaction Chemistry & Engineering (2023).

2. Grzybowski, Bartosz A., et al. "The'wired'universe of organic chemistry." Nature Chemistry 1.1 (2009): 31-36.

3. Fialkowski, Marcin, et al. "Architecture and evolution of organic chemistry." Angewandte Chemie International Edition 44.44 (2005): 7263-7269.

4. Jacob, Philipp-Maximilian, and Alexei Lapkin. "Statistics of the network of organic chemistry." Reaction Chemistry & Engineering 3.1 (2018): 102-118.