TopHap

TopHap infers key phylogenetic structures from large genome collections characterized by limited diversity by focusing on spatiotemporally common haplotypes to improve phylogenetic reconstruction.

Key Features:

• Haplotype identification: Determines spatiotemporally common haplotypes by identifying positions harboring common variants.
• Efficient phylogenetic inference: Enhances signal-to-noise ratio by focusing on common haplotypes of common variants, enabling more accurate trees when few phylogenetically informative sites exist.
• Bootstrap resampling strategy: Employs a novel spatiotemporal bootstrap that resamples genomes across space and time to assess topological robustness.
• Computational efficiency: Reduces computational time compared to conventional phylogenetic methods to enable analysis of very large datasets.
• Robustness to error and recurrent mutation: Mitigates the effects of sequencing errors and recurrent or backward mutations that can obscure true evolutionary relationships.

Scientific Applications:

• SARS-CoV-2 68KG phylogeny: Constructed evolutionary trees for major SARS-CoV-2 haplotypes from 68,057 genomes (68KG), producing results consistent with mutation trees from co-occurrence patterns and recovering key phylogenetic relationships.
• Root evaluation and early lineage inference: Used to evaluate alternative roots of the SARS-CoV-2 phylogeny, suggesting early 2019 samples likely evolved through four mutations from a common ancestor.
• Large-scale variant reconstruction: Applied to over one million genomes to reconstruct evolutionary relationships among major variants and to elucidate origins of variants of concern.

Methodology:

Identify positions harboring common variants, determine spatiotemporally common haplotypes from those positions, construct phylogenetic trees based on the resulting haplotypes, and assess tree robustness with spatiotemporal bootstrap resampling of genomes.