A Computational Framework for Naloxone Biosynthetic Pathway Design using Deep Learning

Other Titles


Opioids are a class of drugs highly valued for their potent analgesic properties; however, they are also highly addictive and cause severe side effects, including death because of respiratory depression. The Centers for Disease Control and Prevention estimate that 130 Americans die daily from opioid overdoses and that the number of opioid overdose deaths in 2017 represents a six-fold increase compared to 1999. Alternative manufacturing methods are needed to overcome the cost and availability of naloxone, an opioid overdose antidote. Toward this need, we developed a computational pipeline to design novel biosynthetic routes from morphine to naloxone. This pipeline combined deep learning with computational biochemistry in an in-silico framework to efficiently explore the biosynthetic pathways' design space.In this study, we assembled metabolic reaction and enzymatic template data from public databases. Then we enriched the reaction dataset with artificial metabolic reactions generated by enzymatic templates. Two neural network-based pathway ranking models were trained as binary classifiers, taking potential reactant and product pairs as input and quantifying the likelihood of a 1-step/2-step enzymatic pathway. Combining these two models with enzymatic templates, we built a multistep retrobiosynthesis pipeline and validated it by computationally reproducing natural and non-natural pathways. While the initial pathway prediction results were encouraging, the framework does have its limitations. To enhance our synthetic pathways generation; improving our precursor candidate rankings' accuracy, we have developed an additional model incorporating thermodynamic constraints. This model takes potential reactant and product pairs as input and predicts the feasibility of chemical reactions within a retrosynthetic workflow, leveraging thermodynamic principles. By integrating thermodynamic considerations, we aim to refine further and enhance our pathway predictions' accuracy, which will be the groundwork for guiding the experimental synthesis of naloxone in our lab.

Journal / Series

Volume & Issue


82 pages


Date Issued




Convolutional Neural Networks; Deep learning model; Network Generation; Retrobiosynthesis


Effective Date

Expiration Date




Union Local


Number of Workers

Committee Chair

Varner, Jeffrey

Committee Co-Chair

Committee Member

You, Fengqi

Degree Discipline

Chemical Engineering

Degree Name

M.S., Chemical Engineering

Degree Level

Master of Science

Related Version

Related DOI

Related To

Related Part

Based on Related Item

Has Other Format(s)

Part of Related Item

Related To

Related Publication(s)

Link(s) to Related Publication(s)


Link(s) to Reference(s)

Previously Published As

Government Document




Other Identifiers


Attribution 4.0 International


dissertation or thesis

Accessibility Feature

Accessibility Hazard

Accessibility Summary

Link(s) to Catalog Record