Is it possible to predict a pathogen’s next move? Novel tool for detecting pathogen fitness says yes!

By Dr. Chinta SidharthanReviewed by Lily Ramsey, LLMJan 13 2025

Pathogen evolution is a central theme in disease ecology, as genetic diversity and fitness changes in pathogens have a significant impact on disease transmission, immune evasion, and public health outcomes. Understanding these dynamics is crucial for managing infectious diseases effectively.

In a recent study published in Nature, researchers developed the concept of “phylowave” — a scalable and innovative method to analyze phylogenetic trees for identifying pathogen lineages with increased fitness.

By applying this approach to pathogens such as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), Bordetella pertussis, and others, the researchers demonstrated the ability of the method to detect known and previously undetected lineages.

​​​​​​​Study: Learning the fitness dynamics of pathogens from phylogenies. Image Credit: HTWE/Shutterstock.com

Background

Pathogen evolution is constant and strongly driven by factors such as immune evasion and environmental changes, leading to the emergence of genetically diverse lineages.

Identifying lineages with increased fitness is vital for humans to predict and manage outbreaks. However, research on pathogen evolution is often restricted to heavily studied systems such as SARS-CoV-2.

Furthermore, existing methods rely on predefined lineage classifications, which limits their ability to detect emergent lineages objectively. Phylogenetic approaches offer an alternative by examining branching patterns to infer fitness but are computationally intensive and sensitive to sampling biases.

While some models can track how a pathogen population is organized or how its size changes over time, they cannot always connect specific genetic changes to how well a pathogen spreads or survives. Current challenges also include defining fitness-linked genetic changes and addressing the biases in sequence data collection.

The Current Study

In the present study, researchers from European universities introduced the concept of “phylowave”, a method to assess pathogen fitness by analyzing phylogenetic trees. They utilized genetic datasets from pathogens such as SARS-CoV-2, influenza A hemagglutinin 3 neuraminidase 2 (H3N2), Bordetella pertussis, and Mycobacterium tuberculosis.

Data for SARS-CoV-2 and H3N2 were obtained from the open-source project NextStrain, while B. pertussis and M. tuberculosis sequences were sourced from national and regional datasets.

Phylowave was applied to calculate an index for each node in the phylogenetic tree based on pairwise genetic distances within a specific timeframe. This index tracked population composition changes and focused on recent evolutionary dynamics.

To compute the index, the researchers adjusted for timescales and window widths specific to each pathogen, ensuring robustness to sampling biases.

They then applied a multinomial logistic model to quantify fitness, assuming constant relative growth rates. The performance of the method was validated through simulations mimicking varying fitness advantages and sampling intensities.

Phylowave was further benchmarked against existing lineage identification methods. The researchers reconstructed phylogenies for B. pertussis using time-resolved single nucleotide polymorphism (SNP) alignments and Bayesian models. For SARS-CoV-2 and H3N2, they used pre-computed phylogenies.

Moreover, mutations defining lineages were identified based on their prevalence and absence in ancestral populations and the functional relevance of the mutations. The researchers designed this approach to minimize computational complexity while ensuring sensitivity to emergent lineages.

Additionally, the study included robustness tests to assess the reliability of the method under different sampling scenarios, while real-time applicability was assessed by truncating SARS-CoV-2 datasets and simulating early detection of new lineages.

Major Findings

The researchers demonstrated that phylowave could reliably detect lineages with fitness differences across diverse pathogens. For SARS-CoV-2, it identified variants such as Omicron sub-lineages with distinct fitness advantages.

These lineages showed dynamic changes in population prevalence, which were correlated to known transmissibility increases. Similarly, the method also captured the evolutionary dynamics of influenza A H3N2, matching established global clades while highlighting persistent lineages averaging 3.9 years of dominance.

In the case of B. pertussis, phylowave uncovered three previously unrecognized lineages that emerged after vaccine shifts in France, demonstrating how immune pressures drive fitness changes.

These lineages exhibited the highest fitness among sampled strains, emphasizing the impact of vaccination on pathogen evolution. Similarly, M. tuberculosis analysis revealed two rising lineages in Russia that were linked to antimicrobial resistance, with stable overall fitness reflecting the long-standing diversity of the pathogen population.

The method also identified lineage-defining mutations, especially in genes related to antigenic sites, virulence, and drug resistance. For SARS-CoV-2, significant mutations were found to be clustered in the receptor-binding domain of the spike protein.

H3N2 mutations were concentrated in the antigenic regions of hemagglutinin, confirming their relevance to immune evasion. In B. pertussis, mutations were found in genes encoding vaccine antigens, which indicated potential immune-driven selection.

Conclusions

Overall, the study demonstrated that phylowave was a novel and robust method for tracking pathogen fitness through phylogenetic analysis. By identifying lineages with differential fitness across multiple pathogens, it exhibited its utility in understanding evolutionary pressures.

Phylowave’s ability to detect genetic changes linked to fitness shifts has significant implications for public health and in enabling timely interventions.

Furthermore, its adaptability across pathogens makes it a valuable tool for real-time monitoring and evolutionary research.