Phylogenetic Analysis of the Third Wave Genome Genome Germany
Over 4750 samples were sequenced between November 2020 and March 2022; of these genomes, 3493 fulfilled quality criteria for phylogenetic and spatiotemporal analyses.
Phylogenetic analysis showed that ZJ11 genotype caused most human infections during wave V. Assumption-free k-nearest neighbour clustering at county level revealed groups of German states with similar spreading dynamics; derivative plots of weekly incidence rates depict an exponential temporal progression up until its partial shutdown in October.
Molecular Characterization
Scientists have long requested increased funding for functional genomics research. Now Germany is responding by investing 350 million DM to stimulate study of functional genomics. The money will create a national network of researchers focused on understanding molecular mechanisms of disease while also encouraging collaboration across borders and international organizations. Unfortunately, some scientists worry that its three year time frame and lack of guaranteed research funding may limit its impact.
Molecular surveillance of SARS-CoV-2 outbreaks is critical to understanding their progression, yet global or even national genome sequencing only provides limited insight. To provide more accurate data, CoMV-Gen, launched between November and December of 2020 as a regional molecular surveillance network encompassing laboratories from Mecklenburg-Western Pomerania (M-V), is a regional molecular surveillance network consisting of nearly equal sample collections across each county in M-V whose samples were sequenced using Illumina and Nanopore technologies to identify genotypes present within circulation.
This project seeks to identify specific mutations that contribute to viral transmission and pathogenesis. More specifically, our team will identify amino acid changes in spike proteins that correlate with increased transmissibility or mortality rates; and use genomics and phylogenetics techniques to monitor their spread through populations.
To conduct molecular characterization, the team employed computational models of interactions between spike protein and its acetylase. Molecular dynamics simulations were then run to compare interaction profiles between wild-type and Macae-mutated spike protein residues. Interactions between proteins were measured as the percentage of simulation time that their residues made contact. This allowed scientists to isolate amino acids contributing to high and low interactions between acetylase and spike proteins. Interactions between virus particles can influence their ability to bind receptors on cells and initiate infection. As such, the results of this research could inform future experimental designs for SARS-CoV-2 vaccines as well as identify possible targets for antiviral therapy drugs.
Phylogenetics
Phylogenetics is the study of tree-like structures such as biological species or virus populations based on evolutionary relationships. By conducting analyses on sampled genomes, phylogenetic analyses provide invaluable information about them, such as their history, growth rate and reproduction number, outbreak source and transmission chain dynamics as well as potential interactions between patient treatment plans and infection dynamics in host hosts.
Phylogenetic analysis played a vital role in global efforts to understand and respond to the COVID-19 pandemic during its first year, when thousands of full or partial severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) genomes were sequenced and made public1,2,3,4.
SARS-CoV-2 is an orthomyxovirus that causes a potentially lethal illness in humans. The virus spreads via droplets or contact with surfaces contaminated with saliva, mucus or urine and leads to symptoms including fever, respiratory distress and gastrointestinal disturbance.
As SARS-CoV-2 spread globally in early 2020, phylogenetic analyses identified and classified outbreaks, described transmission chains, estimated the time to most recent common ancestor (TMRCA), and estimated its genetic distance5.6. Additionally, phylodynamic methods – which take population movement into account as well as lags between patients’ exposure – were utilized to estimate international virus spread rates, identify clusters and outbreak sources as well as quantify patient infectiousness7,8,9.
An evolutionary perspective can also assist with pandemic interventions. Non-pharmaceutical measures, such as travel restrictions and person-to-person distancing can break transmission chains; and phylodynamic analyses can assess their effectiveness by identifying factors affecting intervention success8.
Tracking VOCs (genetic variants with mutations that increase transmissibility or immunity evasion) using phylodynamic methods9 has allowed scientists to monitor their development. Variants with higher mutation frequencies tend to have greater transmission potential; however, other factors, including founder effects, gene linkage and ascertainment bias11 can influence this assessment. Phylodynamic models can therefore provide a useful means of quantifying the effect of interventions on viral transmission, and highlighting key areas for further study. Notably, current research recruitment, evaluation and funding systems disincentivize long-term involvement of researchers skilled in phylodynamic analysis in public health surveillance and control activities12. Therefore, new career pathways and evaluation systems designed to promote such interactions must be created in order to encourage them.
Spatio-Temporal Analyses
Analysis of genome sequences gathered across Germany of SARS-CoV-2 samples has revealed spatial variations in the third pandemic wave’s dynamics, using phylogeographic methods in combination with mathematical time series and pattern recognition analysis techniques.
Results revealed that most imported SARS-CoV-2 lineages originated during a two-month window prior and post Christmas holidays (Fig. 1B), suggesting holiday-related travel played an integral role in both importing new lineages into Germany as well as spreading virus within German samplings consistently across time and space.
Its effects were evident in high incidence rates both before and after the holidays, with localized exponential growth patterns across Germany. While partial shutdown measures in November had some effects, near-linear growth occurred in some areas despite partial shutdown actions taken, which was eventually corrected with additional measures implemented in December.
Spatiotemporal dynamics were further affected by national and state-specific nonpharmaceutical interventions (NPI). We measured their impact using an approach that combined mathematical methods for velocity measurement with clustering of regions based on spreading dynamics to identify any regional features not easily explained by age or other factors that might explain them.
Our findings demonstrated that both federal and state authorities implemented many NPIs that reduced both SARS-CoV-2 importations rates as well as its speed of within-country spread, most effectively the nationwide vaccination programme and restrictions of public movements and gatherings.
An interesting correlation was also discovered between lineage transmissions and state populations, as measured by Pearson correlation coefficient (CC), 0.86. This finding supported our hypothesis that larger population sizes lead to faster internal spread of SARS-CoV-2.
North Rhine-Westphalia, Bavaria and Baden-Wurttemberg had the highest correlation coefficient values among population-rich states (Fig. S3); these correlations reflected both lineage transmissions within these states as well as overall confirmed cases. Conversely, Schleswig-Holstein had lower connectivity with other states resulting in slower SARS-CoV-2 intrastate transmission rates.
Variation in Mutational Dynamics
We examined the mutational dynamics of circulating SARS-CoV-2 variants and discovered their transmission patterns were highly dynamic. A wide diversity of SARS-CoV-2 genotypes could be found, particularly among Delta and Omicron clades, with frequent switching between them, often associated with substantial variation in mutational load within each one. Our analyses also showed evidence for synergic and epistatic interactions that could contribute to rapid spread.
Time series analyses were performed on SARS-CoV-2 mutations for 16 months from November 2020-March 2022.2 Variants that predominated in M-V during this time were identified based on their characteristic mutational signature – such as number and distribution of their defining mutations – while frequency measurements and analyses were measured and performed both monthly and weekly, with heatmaps, cumulative frequency graphs, or frequency frequency graphs used as visualization techniques.
Normalized Total Mutal Load (TML) values were consistently higher for Gamma, Alpha, Lambda, and B.1.1.348 than all other samples (Fig. 1a). This differential in mutational load could be explained by their viruses’ higher Spike mutational load characteristic of these strains (see below).
Distribution of Mutational Load Within Samples The distribution of mutational load was variable among samples, suggesting a heterogeneous genetic background. Prominent mutations included 241:C>A in upstream 5′-UTR region, silent mutation F924F in nsp3 gene and missense mutation P4715L in nsp12. Omicron Clades Display more Variability with their Characteristic Mutations As they comprised several recurrent mutations such as T19R, G142D, 156_158del, and S194L mutations as well as stop gain mutations Y71Y and E39X in membrane protein genes.
M-V mirrored global trends closely, but with one month lag. Omicron clade members were able to rapidly replace Delta variants more rapidly than in Europe or elsewhere, providing us with evidence that SARS-CoV-2 genotypes possess unique mutational landscape and genetic backgrounds which contribute significantly to viral transmission and spread; furthermore, our study of M-V pandemic wave provides us with a baseline for future studies evaluating vaccination’s effect in real world settings.
