Abstract
Exploration of computational approaches for including metabolism information in machine learning models for toxicity prediction.
Highlights
The metabolic system has evolved as the primary defense system against xenobiotic, potentially toxic substances
By comparing the molecular properties of the parent compounds and their predicted metabolites (Fig. 3 reports on the AMES and micronucleus test (MNT) data sets; the graphs for the other endpoints are provided in Fig. S1†) we found the latter to have, averaged over all endpoints, a higher molecular weight (+43.9 Da) as well as a larger polar surface area (+44.4 A2)
The predicted metabolites tended to have a lower log P value than the parent compounds (À1.5; averaged over all endpoints). These shi s are primarily a result of the addition of polar groups to the parent compounds, which make them more water soluble and easier to excrete. This observation is in concordance with the higher number of hydrogen bond donors and acceptors observed in metabolites compared to parent compounds (1.8 more hydrogen bond donors and acceptors on average; Fig. 3)
Summary
The metabolic system has evolved as the primary defense system against xenobiotic, potentially toxic substances. A minority of metabolites produced by the metabolic system are more active. Dmitriev et al.[2] built linear models for the prediction of rat acute toxicity using self-consistent regression, thereby considering parent compounds and measured metabolites. More speci cally, they trained a model on about 3000 parent compounds and used it to predict the LD50 value of 37 test parent compounds and their measured metabolites (around 200 known metabolites). Only minor improvements in the overall performance of the model were achieved compared to using only the predicted probability of the parent compounds (R2 increased from 0.78 to 0.81 and RMSE remained at 0.49).
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
