Effectiveness of molecular fingerprints for exploring the chemical space of natural products

Natural products are a diverse class of compounds with promising biological properties, such as high potency and excellent selectivity. However, they have different structural motifs than typical drug-like compounds, e.g., a wider range of molecular weight, multiple stereocenters and higher fraction of sp3-hybridized carbons. This makes the encoding of natural products via molecular fingerprints difficult, thus restricting their use in cheminformatics studies. To tackle this issue, we reinvestigated over 30 years of research to systematically evaluate which molecular fingerprint provides the best performance on the natural product chemical space. We considered 20 molecular fingerprints from four different sources, which we then benchmarked on over 100000 unique natural products from the COCONUT database. Our results show that despite their apparent comparability, different fingerprinting algorithms can provide fundamentally different views of the natural product chemical space, leading to substantial differences in pairwise similarity. Surprisingly, circular fingerprints, despite being the most popular molecular fingerprint class, were outperformed by path-based, substructure-based and string-based encodings for supervised classification modeling of natural products, highlighting the need to evaluate multiple fingerprinting algorithms for optimal performance and suggesting new areas of further research. Finally, we provide an open-source Python package for computing all the 20 molecular fingerprints considered in the study, as well as data and scripts necessary to reproduce the results, at https://github.com/dahvida/NP_Fingerprints.