A QSP approach based on a graphical model to improve mRNA vaccine platforms
Analyzing and integrating multi-source knowledge and data, with the goal of clarifying the intricate biological processes involved in the mechanism of action of mRNA vaccines, and simulating in silico the induced immunogenicity depending on different vaccine doses, delivery systems and animal species.
What we did
A Natural Language Processing (NLP) pipeline for automated knowledge extraction was developed to gather biological evidence about mRNA vaccines. The retrieved information was merged into an interactive mechanistic graphical model summarizing the chain of events in terms of immunogenicity occurring after administering mRNA vaccines to mice and cynomolgus monkeys. We extended the mechanistic graphical model to better characterize the early events after mRNA vaccine injection through a QSP model, which supports the mRNA vaccine development by providing the chance to assess multiple optimization strategies in silico.
Based on our literature mining NLP pipeline we realized a highly customizable tool that led to the mechanistic graphical model, a sort of road map that we propose to assess the current knowledge and uncover the blind spots in the human immune response to mRNA vaccines.
Leonardelli L, Lofano G, Selvaggio G, Parolo S, Giampiccolo S, Tomasoni D, Domenici E, Priami C, Song H, Medini D, Marchetti L, Siena E. Literature Mining and Mechanistic Graphical Modelling to Improve mRNA Vaccine Platforms. Front Immunol. 2021 Sep 7;12:738388. https://doi.org/10.3389/fimmu.2021.738388.
Selvaggio G, Leonardelli L, Lofano G, Fresnay S, Parolo S, Medini D, Siena E, Marchetti L. A quantitative systems pharmacology approach to support mRNA vaccine development and optimization. CPT Pharmacometrics Syst Pharmacol. 2021 Oct 21. https://doi.org/10.1002/psp4.12721.