European Conference on Mathematical and Theoretical Biology (ECMTB 2026)

Granger-causality analysis reveals defective viral genomes with antiviral potential from longitudinal infection data

16 Jul 2026, 18:00

20m

03.01 - SR (University of Graz)

Contributed Talk Immunobiology & Infection Contributed Talks

Speaker

Tanja Laske (Leibniz Institute of Virology, Viral Systems Modeling, Hamburg, Germany)

Description

Defective viral genomes (DVGs) interfere with infectious standard virus (STV) replication and are considered promising antiviral agents. In longitudinal influenza A virus (IAV) infections, DVG accumulation drives oscillatory virus dynamics (von-Magnus effect [1]). Experimental evaluation of individual DVGs is resource-intensive, motivating computational prioritization.
We aim to identify DVGs influencing STV titers using Granger-causality analysis of sequencing data from a longitudinal IAV infection [2].
For 1,968 DVGs, two ordinary least squares models were compared: a restricted model trained on STV titers, and a full model including DVG trajectories. Upon evaluation of goodness of predictions, DVGs were classified as Granger-causing (DVG predicts STV titers), Granger-caused (STV titers predict DVG), Granger-bi-directional (mutual predictive influence), or non-related [3,4].
We found that 109 DVGs significantly influenced STV titers. A previously validated DVG that reduced STV titers by five orders of magnitude in experiments was classified as Granger-bi-directional, whereas a candidate reducing STV titers by only three orders of magnitude was categorized as non-related. These results indicate that Granger causality-based classification reflects antiviral efficacy and enables prioritization of potent DVGs.
This proof-of-concept shows that Granger-causality allows systematic identification of DVGs with antiviral potential, reducing experimental effort.

Bibliography

1. Von Magnus P. Incomplete forms of influenza virus. Adv Virus Res. 1954;2: 59–79. doi:10.1016/s0065-3527(08)60529-1

2. Pelz L, Rüdiger D, Dogra T, Alnaji FG, Genzel Y, Brooke CB, et al. Semi-continuous Propagation of Influenza A Virus and Its Defective Interfering Particles: Analyzing the Dynamic Competition To Select Candidates for Antiviral Therapy. J Virol. 2021;95: e0117421. doi:10.1128/JVI.01174-21

3. Granger CWJ. Investigating Causal Relations by Econometric Models and Cross-spectral Methods. Econometrica. 1969;37: 424. doi:10.2307/1912791

4. Lütkepohl H. New introduction to multiple time series analysis. Berlin: Springer; 2005.