.jpg)
String representations like SMILES are among the most popular and convenient for encoding molecules in machine learning models. Graphs are inherent in SMILES, making them highly expressive and flexible in the generation of other representations, including fingerprints for the computation of molecular similarity. However, the complex nature of chemical bonding between metals and ligands has hampered the adoption of SMILES in the field of metal-organic chemistry since the chemoinformatics term appeared in the late 90s. The present work explores three different approaches to the formulation of SMILES strings for metal complexes based on 1) extended Hückel calculations, 2) natural bond orbital analysis, and 3) strings from the Cambridge Structural Database.
One of the contributors from the Jensen group in the University of Copenhagen, Magnus Strandgaard, visited the Hylleraas Centre in a 3-month PhD exchange in 2023.
Abstract
We present a method for creating RDKit-parsable SMILES for transition metal complexes (TMCs) based on xyz-coordinates and overall charge of the complex. This can be viewed as an extension to the program xyz2mol that does the same for organic molecules. The only dependency is RDKit, which makes it widely applicable. One thing that has been lacking when it comes to generating SMILES from structure for TMCs is an existing SMILES dataset to compare with. Therefore, we also generate SMILES two other ways; one where ligand charges and TMC connectivity are based on natural bond orbital (NBO) analysis from density functional theory (DFT) calculations utilizing recent work by Kneiding et al. (Digit Discov 2: 618–633, 2023). Another one fixes SMILES available through the Cambridge Structural Database (CSD), making them parsable by RDKit. We compare these three different ways of obtaining SMILES for a subset of the CSD (tmQMg) and find >70% agreement for all three pairs. We utilize these SMILES to make simple molecular fingerprint (FP) and graph-based representations of the molecules to be used in the context of machine learning. Comparing with the graphs made by Kneiding et al. where nodes and edges are featurized with DFT properties, we find that depending on the target property (polarizability, HOMO-LUMO gap or dipole moment) the SMILES based representations can perform equally well. This makes them very suitable as baseline-models. Finally we present a dataset of 227k RDKit parsable SMILES for mononuclear TMCs in the CSD.
Reference
Rasmussen, M. H.; Strandgaard, M.; Seumer, J.; Hemmingsen, L. K.; Frei, A.; Balcells, D.; Jensen, J. H. SMILES all around: Structure to SMILES conversion for transition metal complexes J. Cheminf. 2025, 17, 63.
First published online: 28 April 2025.