TempSnap-Trace: A temporal snapshot-based framework for haplotype network tracing

Area of Science:

• Virology
• Computational Biology
• Genomics

Background:

• Tracking large-scale viral evolutionary history presents significant computational challenges.
• Existing methods may struggle with the scale and complexity of viral genomic data.

Purpose of the Study:

• Introduce TempSnap-Trace, a novel computational framework for tracking large-scale viral evolutionary history.
• Enhance viral surveillance and variant origin tracing capabilities.

Main Methods:

• Generate variant-featured haplotype strings from genomic data.
• Employ minimum-cost arborescence network (McAN) algorithm for phylogenetic inference.
• Construct weighted temporal snapshot networks and identify core evolutionary nodes via community detection.

Main Results:

• The framework integrates mutation sites, network topology, and directional information to reconstruct evolutionary pathways.
• Achieved 17.4% increased modularity and 30.1% reduced code length compared to unweighted graph methods.
• Successfully traced SARS-CoV-2 and Mpox variants, identified transmission hubs, and demonstrated a 27-fold speedup over VENAS, processing data that VENAS failed on due to memory limits.

Conclusions:

• TempSnap-Trace is computationally efficient and scalable for large-scale viral surveillance.
• The framework provides distinct advantages for cross-border transmission warning and variant origin tracing.
• Validated utility on SARS-CoV-2 and Mpox, demonstrating robust evolutionary path reconstruction.