XenoSite: Accurately Predicting CYP-Mediated Sites of Metabolism with Neural Networks

Zaretzki, Jed; Matlock, Matthew K.; Swamidass, S. Joshua

doi:10.1021/ci400518g

Cited by 192 publications

(253 citation statements)

References 21 publications

Supporting

Mentioning

246

Contrasting

Order By: Relevance

“…The bulk of our descriptors have been previously shown to be useful for the XenoSite metabolism model, although we supplement them with new reactivity descriptors in the current study. 12 Several of these descriptors have been proposed as reactivity indices, such as the energies of the lowest unoccupied and highest occupied molecular orbitals ( E LUMO and E HOMO ), the maximum nucleophilic and electrophilic delocalizabilities (max[ D N ( r )] and max[ D E ( r )]), and the maximum self-polarizability (max[ π S ( r )]). Moreover, D N ( r ), D E ( r ), and π S ( r ) have been proposed as atom-level reactivity indices that may predict sites of GSH reactivity.…”

Section: Methodsmentioning

confidence: 99%

“…32 We found success with this approach when predicting P450 metabolism. 12 When applied in this study, the strategy was capable of encoding nonlinear relationships and simultaneously making SOR predictions for each atom in a molecule along with GSH reactivity predictions for the molecule as a whole. The validation of those models demonstrated the ability to model effectively GSH reactivity of diverse chemicals, both identifying reactive molecules and sites of reactivity.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Site of Reactivity Models Predict Molecular Reactivity of Diverse Chemicals with Glutathione

Hughes

Miller

Swamidass

2015

Chem. Res. Toxicol.

Self Cite

130

View full text Add to dashboard Cite

Drug toxicity is often caused by electrophilic reactive metabolites that covalently bind to proteins. Consequently, the quantitative strength of a molecule’s reactivity with glutathione (GSH) is a frequently used indicator of its toxicity. Through cysteine, GSH (and proteins) scavenges reactive molecules to form conjugates in the body. GSH conjugates to specific atoms in reactive molecules: their sites of reactivity. The value of knowing a molecule’s sites of reactivity is unexplored in the literature. This study tests the value of site of reactivity data that identifies the atoms within 1213 reactive molecules that conjugate to GSH and builds models to predict molecular reactivity with glutathione. An algorithm originally written to model sites of cytochrome P450 metabolism (called XenoSite) finds clear patterns in molecular structure that identify sites of reactivity within reactive molecules with 90.8% accuracy and separate reactive and unreactive molecules with 80.6% accuracy. Furthermore, the model output strongly correlates with quantitative GSH reactivity data in chemically diverse, external data sets. Site of reactivity data is nearly unstudied in the literature prior to our efforts, yet it contains a strong signal for reactivity that can be utilized to more accurately predict molecule reactivity and, eventually, toxicity.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Site of Reactivity Models Predict Molecular Reactivity of Diverse Chemicals with Glutathione

Hughes

Miller

Swamidass

2015

Chem. Res. Toxicol.

Self Cite

130

View full text Add to dashboard Cite

show abstract

“…An ideal algorithm will rank a sufficient number of active compounds before the inactives, but the rankings of actives relative to other actives and inactives are less important [384]. Computational modeling also has the potential to predict ADMET traits for lead generation [385] and how drugs are metabolized [386].…”

Section: Ligand-based Prediction Of Bioactivitymentioning

confidence: 99%

“…A study of 22 ADMET tasks demonstrated that there are limitations to multi-task transfer learning that are in part a consequence of the degree to which tasks are related [385].…”

Section: Ligand-based Prediction Of Bioactivitymentioning

confidence: 99%

See 1 more Smart Citation

Opportunities and obstacles for deep learning in biology and medicine

Ching

Himmelstein

Beaulieu‐Jones

et al. 2017

Preprint

Self Cite

236

248

View full text Add to dashboard Cite

Deep learning, which describes a class of machine learning algorithms, has recently showed impressive results across a variety of domains. Biology and medicine are data rich, but the data are complex and often ill-understood. Problems of this nature may be particularly well-suited to deep learning techniques. We examine applications of deep learning to a variety of biomedical problems—patient classification, fundamental biological processes, and treatment of patients—and discuss whether deep learning will transform these tasks or if the biomedical sphere poses unique challenges. We find that deep learning has yet to revolutionize or definitively resolve any of these problems, but promising advances have been made on the prior state of the art. Even when improvement over a previous baseline has been modest, we have seen signs that deep learning methods may speed or aid human investigation. More work is needed to address concerns related to interpretability and how to best model each problem. Furthermore, the limited amount of labeled data for training presents problems in some domains, as do legal and privacy constraints on work with sensitive health records. Nonetheless, we foresee deep learning powering changes at both bench and bedside with the potential to transform several areas of biology and medicine.

show abstract

Comprehensive insights into the formation of metabolites of the ghrelin mimetics capromorelin, macimorelin and tabimorelin as potential markers for doping control purposes

Lange

Thomas

Görgens

et al. 2021

Biomedical Chromatography

View full text Add to dashboard Cite

Analytical methods to determine the potential misuse of the ghrelin mimetics capromorelin (CP‐424,391), macimorelin (macrilen, EP‐01572) and tabimorelin (NN703) in sports were developed. Therefore, different extraction strategies, i.e. solid‐phase extraction, protein precipitation, as well as a “dilute‐and‐inject” approach, from urine and EDTA–plasma were assessed and comprehensive in vitro/in vivo experiments were conducted, enabling the identification of reliable target analytes by means of high resolution mass spectrometry. The drugs’ biotransformation led to the preliminary identification of 51 metabolites of capromorelin, 12 metabolites of macimorelin and 13 metabolites of tabimorelin. Seven major metabolites detected in rat urine samples collected post‐administration of 0.5–1.0 mg of a single oral dose underwent in‐depth characterization, facilitating their implementation into future confirmatory test methods. In particular, two macimorelin metabolites exhibiting considerable abundances in post‐administration rat urine samples were detected, which might contribute to an improved sensitivity, specificity, and detection window in case of human sports drug testing programs. Further, the intact drugs were implemented into World Anti‐Doping Agency‐compliant initial testing (limits of detection 0.02–0.60 ng/ml) and confirmation procedures (limits of identification 0.18–0.89 ng/ml) for human urine and blood matrices. The obtained results allow extension of the test spectrum of doping agents in multitarget screening assays for growth hormone‐releasing factors from human urine.

show abstract

XenoSite: Accurately Predicting CYP-Mediated Sites of Metabolism with Neural Networks

Cited by 192 publications

References 21 publications

Site of Reactivity Models Predict Molecular Reactivity of Diverse Chemicals with Glutathione

Site of Reactivity Models Predict Molecular Reactivity of Diverse Chemicals with Glutathione

Opportunities and obstacles for deep learning in biology and medicine

Comprehensive insights into the formation of metabolites of the ghrelin mimetics capromorelin, macimorelin and tabimorelin as potential markers for doping control purposes

Contact Info

Product

Resources

About