Traditional approaches in natural products drug discovery have been highly effective in the discovery of clinically useful agents, accounting by some estimates for over 50% of our approved drugs, either directly or by inspiring related agents to be chemically synthesized.  Nevertheless, the field continues to pioneer new advances in terms of approaches and classes of organisms under study, striving to maintain this record of success while at the same time introducing new innovations. In this regard, two recently developed methods for the automatic classification of natural product structures will be presented.  The first is the established Global Natural Products Social (GNPS) Molecular Networks which is based on mass spectral fragmentation patterns.  This is especially useful for rapid dereplication of known compounds and identification of analogs in a given desired structure class.  Complementing this is the newly introduced Small Molecule Accurate Recognition Technology (SMART) which is based on a deep convolutional neural network applied to <sup>1</sup>H-<sup>13</sup>C HSQC data.  Application of these methodologies is accelerating the pace of natural products identification and structure determination, as will be illustrated by specific examples in the presentation.

New methods for the automated structural classification of natural products

2019 Fall National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.