NPClassifier: A Deep Neural Network-Based Structural Classification Tool for Natural Products

Abstract

Computational approaches such as genome and metabolome mining are becoming essential to natural products (NPs) research. Consequently, a need exists for an automated structure-type classification system to handle the massive amounts of data appearing for NP structures. An ideal semantic ontology for the classification of NPs should go beyond the simple presence/ absence of chemical substructures, but also include the taxonomy of the producing organism, the nature of the biosynthetic pathway, and/or their biological properties. Thus, a holistic and automatic NP classification framework could have considerable value to comprehensively navigate the relatedness of NPs, and especially so when analyzing large numbers of NPs. Here, we introduce NPClassifier, a deep-learning tool for the automated structural classification of NPs from their counted Morgan fingerprints. NPClassifier is expected to accelerate and enhance NP discovery by linking NP structures to their underlying properties.

Feature-Based Molecular Networking—An Exciting Tool to Spot Species of the Genus Cortinarius with Hidden Photosensitizers

“…However, the general occurrence of photoactive pigments, their photopharmaceutical potential, and their ecological role are still rather unclear. In an attempt to explore this phenomenon, this study looked into the chemical space of extracts derived from Cortinarius species representing different phylogenetic lineages for the first time by means of UPLC-HRMS 2 metabolomics tools (i.e., FBMN) [26,72] to facilitate a holistic view of the complex mixture of fungal secondary metabolites. By implementing in vitro data in the generated network, an elegant way to visualize clusters of photoactive features was developed specifically.…”

Multi-omics profiling of Earth’s biomes reveals patterns of diversity and co-occurrence in microbial and metabolite composition across environments

“…Samples were contributed by 34 principal investigators (PIs) of the Earth Microbiome Project 500 (EMP500) Consortium. To achieve more even coverage across microbial environments, we devised an ontology of sample types (i.e., microbial environments), the EMP Ontology (EMPO) (http://www.earthmicrobiome.org/protocols-and-standards/empo/) 19 and selected samples to fill out EMPO categories as broadly as possible. Samples were collected following the EMP sample submission guide 20 .…”

“…In silico structure annotation using structures from biodatabase was done with CSI:FingerID 17 . Systematic class annotations were obtained with CANOPUS 18 and used the NPClassifier ontology 19 . The parameters for SIRIUS tools were set as follows, for SIRIUS: molecular formula candidates retained (80), molecular formula database (ALL), maximum precursor ion m/z computed (750), profile (orbitrap), m/z maximum deviation (10 ppm), ions annotated with MZmine were prioritized and other ions were considered ([M+H3N+H]+, [M+H]+, [M+K]+,[M+Na]+, [M+H-H2O]+, [M+H-H4O2]+, [M+NH4]+); for ZODIAC: the features were split into 10 random subsets for lower computational burden and computed separately with the following parameters: threshold filter (0.9), minimum local connections (0); for CSI:FingerID: m/z maximum deviation (10 ppm) and biological database (BIO).…”

“…Systematic class annotations were obtained with CANOPUS 18 and used the NPClassifier ontology 19 . The parameters for SIRIUS tools were set as follows, for SIRIUS: molecular formula candidates retained (80), molecular formula database (ALL), maximum precursor ion m/z computed (750), profile (orbitrap), m/z maximum deviation (10 ppm), ions annotated with…”

“…University of Massachusetts Amherst, Amherst, Massachusetts, USA 18. Montana State University, Bozeman, Montana, USA 19. University of Arizona, Tucson, Arizona, USA 20.…”