Design of Natural‐Product‐Inspired Multitarget Ligands by Machine Learning

ChemMedChem ◽  
2019 ◽  
Vol 14 (12) ◽  
pp. 1129-1134 ◽  
Author(s):  
Francesca Grisoni ◽  
Daniel Merk ◽  
Lukas Friedrich ◽  
Gisbert Schneider
Keyword(s):  
Machine Learning ◽  
Natural Product

scholarly journals Global analysis of adenylate-forming enzymes reveals β-lactone biosynthesis pathway in pathogenic Nocardia

2020 ◽  
Vol 295 (44) ◽  
pp. 14826-14839
Author(s):  
Serina L. Robinson ◽  
Barbara R. Terlouw ◽  
Megan D. Smith ◽  
Sacha J. Pidot ◽  
Timothy P. Stinear ◽  
...  
Keyword(s):  
Machine Learning ◽  
Natural Product ◽  
Global Analysis ◽  
Gene Clusters ◽  
Divergent Evolution ◽  
Acid Activation ◽  
Biosynthetic Gene ◽  
Biosynthetic Gene Clusters ◽  
Predictive Tool ◽  
Wide Range

Enzymes that cleave ATP to activate carboxylic acids play essential roles in primary and secondary metabolism in all domains of life. Class I adenylate-forming enzymes share a conserved structural fold but act on a wide range of substrates to catalyze reactions involved in bioluminescence, nonribosomal peptide biosynthesis, fatty acid activation, and β-lactone formation. Despite their metabolic importance, the substrates and functions of the vast majority of adenylate-forming enzymes are unknown without tools available to accurately predict them. Given the crucial roles of adenylate-forming enzymes in biosynthesis, this also severely limits our ability to predict natural product structures from biosynthetic gene clusters. Here we used machine learning to predict adenylate-forming enzyme function and substrate specificity from protein sequences. We built a web-based predictive tool and used it to comprehensively map the biochemical diversity of adenylate-forming enzymes across >50,000 candidate biosynthetic gene clusters in bacterial, fungal, and plant genomes. Ancestral phylogenetic reconstruction and sequence similarity networking of enzymes from these clusters suggested divergent evolution of the adenylate-forming superfamily from a core enzyme scaffold most related to contemporary CoA ligases toward more specialized functions including β-lactone synthetases. Our classifier predicted β-lactone synthetases in uncharacterized biosynthetic gene clusters conserved in >90 different strains of Nocardia. To test our prediction, we purified a candidate β-lactone synthetase from Nocardia brasiliensis and reconstituted the biosynthetic pathway in vitro to link the gene cluster to the β-lactone natural product, nocardiolactone. We anticipate that our machine learning approach will aid in functional classification of enzymes and advance natural product discovery.

De-orphaning the marine natural product (±)-marinopyrrole A by computational target prediction and biochemical validation

2017 ◽  
Vol 53 (14) ◽  
pp. 2272-2274 ◽  
Author(s):  
P. Schneider ◽  
G. Schneider
Keyword(s):  
Machine Learning ◽  
Natural Product ◽  
Target Prediction ◽  
Marine Natural Product ◽  
Machine Learning Method ◽  
Learning Method ◽  
Biochemical Validation ◽  
Anticancer Compound

A machine-learning method led to the discovery of the macromolecular targets of the natural anticancer compound marinopyrrol A.

Prediction of Natural Product Classes Using Machine Learning and 13C NMR Spectroscopic Data

2020 ◽  
Vol 60 (7) ◽  
pp. 3376-3386 ◽  
Author(s):  
Saúl H. Martínez-Treviño ◽  
Víctor Uc-Cetina ◽  
María A. Fernández-Herrera ◽  
Gabriel Merino
Keyword(s):  
Machine Learning ◽  
Natural Product ◽  
13C Nmr ◽  
Spectroscopic Data

scholarly journals Focused natural product elucidation by prioritizing high-throughput metabolomic studies with machine learning

2019 ◽  
Author(s):  
Nicholas J. Tobias ◽  
César Parra-Rojas ◽  
Yan-Ni Shi ◽  
Yi-Ming Shi ◽  
Svenja Simonyi ◽  
...  
Keyword(s):  
Machine Learning ◽  
Natural Product ◽  
High Throughput ◽  
Abiotic Factors ◽  
Future Research ◽  
Natural Product Research ◽  
Strain Collection ◽  
Metabolic Variation ◽  
Metabolic Screening ◽  
Different Strains

AbstractBacteria of the genera Photorhabdus and Xenorhabdus produce a plethora of natural products to support their similar symbiotic lifecycles. For many of these compounds, the specific bioactivities are unknown. One common challenge in natural product research when trying to prioritize research efforts is the rediscovery of identical (or highly similar) compounds from different strains. Linking genome sequence to metabolite production can help in overcoming this problem. However, sequences are typically not available for entire collections of organisms. Here we perform a comprehensive metabolic screening using HPLC-MS data associated with a 114-strain collection (58 Photorhabdus and 56 Xenorhabdus) from across Thailand and explore the metabolic variation among the strains, matched with several abiotic factors. We utilize machine learning in order to rank the importance of individual metabolites in determining all given metadata. With this approach, we were able to prioritize metabolites in the context of natural product investigations, leading to the identification of previously unknown compounds. The top three highest-ranking features were associated with Xenorhabdus and attributed to the same chemical entity, cyclo(tetrahydroxybutyrate). This work addresses the need for prioritization in high-throughput metabolomic studies and demonstrates the viability of such an approach in future research.

Artificial Intelligence Recognizes β-Lapachone as an Allosteric 5- Lipoxygenase Inhibitor

2018 ◽  
Author(s):  
Gonçalo Bernardes ◽  
Tiago Rodrigues ◽  
Markus Werner ◽  
Jakob Roth ◽  
Eduardo H. G. da Cruz ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Machine Learning ◽  
Drug Discovery ◽  
Natural Product ◽  
Systems Pharmacology ◽  
Lipoxygenase Inhibitor ◽  
Machine Learning Methods ◽  
Biochemical Assays ◽  
Viable Approach ◽  
Chemical Matter

<div> <div> <div> <p>Chemical matter with often-discarded moieties entails opportunities for drug discovery. Relying on orthogonal ligand-centric machine learning methods, targets were consensually identified as potential counterparts for the fragment-like natural product β-lapachone. Resorting to a comprehensive range of biophysical and biochemical assays, the natural product was validated as a potent, ligand efficient, allosteric and reversible modulator of 5-lipoxygenase (5-LO). Moreover, we provide a rationale for 5-LO-inhibiting chemotypes inspired in the β-lapachone scaffold through a focused analogue library. This work demonstrates the power of artificial intelligence technologies to deconvolute complex phenotypic readouts of clinically relevant chemical matter, leverage natural product-based drug discovery, as an alternative and/or complement to chemoproteomics and as a viable approach for systems pharmacology studies. </p> </div> </div> </div>

From Machine Learning to Natural Product Derivatives that Selectively Activate Transcription Factor PPARγ

ChemMedChem ◽  
2010 ◽  
Vol 5 (2) ◽  
pp. 191-194 ◽  
Author(s):  
Matthias Rupp ◽  
Timon Schroeter ◽  
Ramona Steri ◽  
Heiko Zettl ◽  
Ewgenij Proschak ◽  
...  
Keyword(s):  
Machine Learning ◽  
Transcription Factor ◽  
Natural Product ◽  
Activate Transcription Factor

scholarly journals NP-Scout: Machine Learning Approach for the Quantification and Visualization of the Natural Product-Likeness of Small Molecules

Biomolecules ◽  
2019 ◽  
Vol 9 (2) ◽  
pp. 43 ◽  
Author(s):  
Ya Chen ◽  
Conrad Stork ◽  
Steffen Hirte ◽  
Johannes Kirchmair
Keyword(s):  
Machine Learning ◽  
Natural Products ◽  
Small Molecules ◽  
Natural Product ◽  
Characteristic Curve ◽  
Correlation Coefficients ◽  
Learning Approach ◽  
Data Sets ◽  
Machine Learning Approach ◽  
Synthetic Molecules

Natural products (NPs) remain the most prolific resource for the development of small-molecule drugs. Here we report a new machine learning approach that allows the identification of natural products with high accuracy. The method also generates similarity maps, which highlight atoms that contribute significantly to the classification of small molecules as a natural product or synthetic molecule. The method can hence be utilized to (i) identify natural products in large molecular libraries, (ii) quantify the natural product-likeness of small molecules, and (iii) visualize atoms in small molecules that are characteristic of natural products or synthetic molecules. The models are based on random forest classifiers trained on data sets consisting of more than 265,000 to 322,000 natural products and synthetic molecules. Two-dimensional molecular descriptors, MACCS keys and Morgan2 fingerprints were explored. On an independent test set the models reached areas under the receiver operating characteristic curve (AUC) of 0.997 and Matthews correlation coefficients (MCCs) of 0.954 and higher. The method was further tested on data from the Dictionary of Natural Products, ChEMBL and other resources. The best-performing models are accessible as a free web service at http://npscout.zbh.uni-hamburg.de/npscout.

scholarly journals Global analysis of adenylate-forming enzymes reveals β-lactone biosynthesis pathway in pathogenic Nocardia

2019 ◽  
Author(s):  
Serina L. Robinson ◽  
Barbara R. Terlouw ◽  
Megan D. Smith ◽  
Sacha J. Pidot ◽  
Tim P. Stinear ◽  
...  
Keyword(s):  
Machine Learning ◽  
Natural Product ◽  
Sequence Similarity ◽  
Global Analysis ◽  
Gene Clusters ◽  
Acid Activation ◽  
Biosynthetic Gene ◽  
Biosynthetic Gene Clusters ◽  
Predictive Tool ◽  
Wide Range

ABSTRACTEnzymes that cleave ATP to activate carboxylic acids play essential roles in primary and secondary metabolism in all domains of life. Class I adenylate-forming enzymes share a conserved structural fold but act on a wide range of substrates to catalyze reactions involved in bioluminescence, nonribosomal peptide biosynthesis, fatty acid activation, and β-lactone formation. Despite their metabolic importance, the substrates and catalytic functions of the vast majority of adenylate-forming enzymes are unknown without tools available to accurately predict them. Given the crucial roles of adenylate-forming enzymes in biosynthesis, this also severely limits our ability to predict natural product structures from biosynthetic gene clusters. Here we used machine learning to predict adenylate-forming enzyme function and substrate specificity from protein sequence. We built a web-based predictive tool and used it to comprehensively map the biochemical diversity of adenylate-forming enzymes across >50,000 candidate biosynthetic gene clusters in bacterial, fungal, and plant genomes. Ancestral enzyme reconstruction and sequence similarity networking revealed a ‘hub’ topology suggesting radial divergence of the adenylate-forming superfamily from a core enzyme scaffold most related to contemporary aryl-CoA ligases. Our classifier also predicted β-lactone synthetases in novel biosynthetic gene clusters conserved across >90 different strains of Nocardia. To test our computational predictions, we purified a candidate β-lactone synthetase from Nocardia brasiliensis and reconstituted the biosynthetic pathway in vitro to link the gene cluster to the β-lactone natural product, nocardiolactone. We anticipate our machine learning approach will aid in functional classification of enzymes and advance natural product discovery.

Sign in / Sign up

Export Citation Format

Share Document