B04: Linking macroscopic evolution with molecular processes for rapidly evolving viral pathogens
Projects of the CRC 1768
B04: Linking macroscopic evolution with molecular processes for rapidly evolving viral pathogens
Viral pathogens such as SARS-CoV-2 and human influenza A viruses are single-stranded RNA viruses with substantial capacity to mutate and to adapt to the human host for more efficient replication and spread. A multitude of factors affect the evolutionary patterns left in their genomes, such as adaptation to changing host immunity or for more efficient replication, phylogenetic spread, as well as uncharacterised processes on the cellular level. Continuous changes in the surface antigens of these viruses allow them to evade host immunity developed through either prior infection from previous strains or from vaccination. This capacity of a virus, known as immune escape, facilitates the reinfection of individuals. Consequently, vaccines protecting against such viruses need to be frequently updated to maintain their effectiveness against circulating variants. We hypothesise that our understanding of the complex interplay of these various processes from large-scale viral genome data can be improved by careful analysis and deconvolution with tailor-made computational techniques. This improved understanding of viral evolution will make it even more predictable on the population level and facilitate the early identification of future emerging, antigenically altered variants of concern for public health.
We have recently developed techniques that allowed us to predict the emergence of relevant variants of SARS-CoV-2, as reported by the World Health Organization (WHO), substantially prior to this classification and to their reaching their maximal abundances. We are also able to identify lineages with substantial antigenic alterations, which can inform considerations regarding vaccine strain updates. In this project, we will combine data-driven analytics of population level viral diversity with molecular modelling across scales to link macroscopic viral evolution on a population level to molecular processes within the cell. By combining data-driven surveillance and simulation, we will be able to study evolutionary and epidemiological phenomena in both data and models. These include: (1) Developing approaches for early detection and further characterisation of antigenically or otherwise phenotypically altered lineages identified by the WHO as Variants of Concern (VOCs) via viral genomic surveillance (G3). Early detection methods for identifying antigenically altered lineages classified by the World Health Organization as concerning, of interest, or under monitoring have recently been developed in the McHardy lab.
We will extend this approach to predict combinations of amino acid changes driving future predominant lineages, enabling earlier detection of potential VOCs than current methods. (2) Developing a multi-scale simulation platform consisting of (i) a micro-level – simulating virus replication within a cell; and (ii) a macro-level – simulating virus evolution. At the micro level, we will develop a new type of rule-based description language, including RNA and dynamical compartments. The rule-based replication cycle model will allow us to trace a mutation through the replication cycle. This trace helps to clarify the putative effects of the mutation on the dynamics of the replication cycle, and also to explain how the mutation could affect the fitness of the virus. (3) Combining the results from all work packages to disentangle the contributions of genetic drift, antigenic drift, and currently uncharacterised processes on the genetic diversity of circulating lineages and to study the evolutionary role of the “not-yet-explained” mutations that influence the replication cycle of the virus within the host cell.
Project Overview
Our project is based on the hypothesis that data-driven population-level evolutionary analytics linked with molecular modelling of virus replication within a host cell will enhance our understanding of the mechanisms of viral evolution and our predictions of future concerning lineages of SARS-CoV-2. This project will provide novel data-driven methods to support scientists in selecting variants for in-depth experimental characterisation and public health efforts such as those of the World Health Organization’s Technical Advisory Group on Viral Evolution. It will generate predictions about novel, emerging variants of interest and their antigenic alterations with potential relevance for vaccine strain updates, as well as identify evolutionary hotspots at the level of protein structures and RNA. Accordingly, our objectives are as follows: (1) to establish a curated database of mutations, in both coding and non-coding regions, annotated with evolutionary fitness advantage estimates derived from diverse data sources and both established and novel methods (WP 1); (2) to design a predictive framework for lineage forecasting that integrates and evaluates genomic features together with experimental data – such as antibody escape assays and cell-entry efficiency measurements – to identify amino acid changes driving lineage success (WP 2); (3) to develop a multi-scale simulation platform consisting of (i) a micro level, describing virus replication within a cell with a new rule-based language including RNA and dynamical compartments (WP 3), and (ii) a macro-level modelling virus evolution (WP 4); and (4) to investigate (in WP 5) novel mechanisms of viral evolution by integrating genomic surveillance data with rule-based modelling of the viral replication cycle, specifically by examining (i) intra-cellular mechanisms that underlie the fixation of specific mutations, (ii) the contributions of genetic drift, antigenic drift, and other uncharacterised processes to the genetic diversity of circulating lineages, and (iii) the evolutionary role of “not-yet-explained mutations” that influence viral replication within the host cell.
Our tools will be initially developed for SARS-CoV-2 and generalised to other viruses in subsequent funding periods. The rule-based modelling language will not contain any SARS-CoV-2 specific elements and will thus be applicable to any virus, already in the first funding phase.
- Tool to be developed: A method and database integrating genomic and experimental data to predict future, emerging antigenic types of circulating lineages to support researchers and public health surveillance efforts; initially focused on SARS-CoV-2, then to be extended to other rapidly evolving RNA viruses.
- Tool to be developed: A rule-based simulation platform to model molecular replication processes of any virus within any host cell, and to link the replication process to virus evolution at the genomic level.
Hypothesis enabled by the proposed tool: Tools that allow improved data-driven population level evolutionary and phytogeographic analytics linked with molecular modelling of virus replication within a host cell will enhance our understanding of the mechanisms of viral evolution and our predictions of future critical viral lineages.
Overarching CRC goals: Our project couples population-scale genomic surveillance with a multi-scale, rule-based replication-cycle simulator to mechanistically link within-cell processes to macroscopic viral evolution and antigenic change (G2). The resulting inference framework and database prospectively predict emerging antigenically altered lineages and key amino-acid combinations (G1), enabling earlier detection and vaccine-strain considerations while yielding transferable principles that generalise constraints on virus-host interaction (G3, G4).
Work Packages (WP):
- WP 1: Database of phenotype-altering and neutral mutations and amino acid changes in viral evolution (McHardy)
- WP 2: Early prediction of predominant, antigenically altered lineage types (McHardy)
- WP 3: Rule-based simulation platform of in-cell replication cycle (Dittrich)
- WP 4: Tools for sequence-level evolutionary and epidemiological dynamics (Dittrich)
- WP 5: Search for novel mechanisms of viral evolution (McHardy/Dittrich)
Team Members
Dr. Mohammad-Hadi Foroughmand-Araabi
PostDoc
PhD B04 1
PhD Student
PhD B04 2
PhD Student
2025
In silico genomic surveillance by CoVerage predicts and characterizes SARS-CoV-2 Variants of Interest Journal Article
In: Nat Commun, vol. 16, iss. 1, pp. 6281, 2025.
2024
Importations of SARS-CoV-2 lineages decline after nonpharmaceutical interventions in phylogeographic analyses Journal Article
In: Nat Commun, vol. 15, no. 1, pp. 5267, 2024.
2021
Hepatitis C reference viruses highlight potent antibody responses and diverse viral functional interactions with neutralising antibodies. Journal Article
In: Gut, vol. 70, iss. 9, pp. 1734–1745, 2021.
Linking network structure and dynamics to describe the set of persistent species in reaction diffusion systems Journal Article
In: SIAM J. Appl. Dyn. Syst., vol. 20, iss. 4, pp. 2037–2076, 2021.
2020
Structure and hierarchy of SARS-CoV-2 infection dynamics models revealed by reaction network analysis Journal Article
In: Viruses, vol. 13, iss. 1, pp. 14, 2020.
2019
Structure and hierarchy of influenza virus models revealed by reaction network analysis. Journal Article
In: Viruses, vol. 11, iss. 5, 2019.
Spatial rule-based simulations: The SRSim software. Journal Article
In: Methods Mol Biol, vol. 1945, pp. 231–249, 2019.
2014
Computational prediction of vaccine strains for human influenza A (H3N2) viruses. Journal Article
In: J Virol, vol. 88, iss. 20, pp. 12123–12132, 2014.
2012
Inference of genotype–phenotype relationships in the antigenic evolution of human influenza A (H3N2) viruses Journal Article
In: PLoS Comput Biol, vol. 8, iss. 4, pp. e1002492, 2012.
2010
Rule-based spatial modeling with diffusing, geometrically constrained molecules Journal Article
In: BMC Bioinformatics, vol. 11, iss. 1, 2010.