International Edition
Latest News
Business

Phylogenomic Analysis of S-Layer Proteins Across the Tree of Life Reveals Anchoring Domain Swapping Obscures Evolutionary Histories

In January 2026, researchers expanded the Tree of Life dataset to 2,705 Bacteria, 315 Archaea, and 149 Eukaryota genomes, incorporating an additional 127 genomes of organisms with available evidence of S-layers and SLLPs, Nature reported. The updated dataset…

Phylogenomic Analysis of S-Layer Proteins Across the Tree of Life Reveals Anchoring Domain Swapping Obscures Evolutionary

In January 2026, researchers expanded the Tree of Life dataset to 2,705 Bacteria, 315 Archaea, and 149 Eukaryota genomes, incorporating an additional 127 genomes of organisms with available evidence of S-layers and SLLPs, Nature reported.

The updated dataset builds on previous phylogenetic inferences by integrating targeted genomic data. Seven conserved single-copy marker proteins were identified using hmmsearch from HMMER3.3, according to the study.

The resulting orthologue sequences were aligned with MAFFT and trimmed with trimAl in gappyout mode. These sequences were then concatenated using the Biopython module Nexus to infer phylogenies in IQ-TREE under the LG + G4 model with 1,000 ultrafast bootstraps.

Because of the large volume of available Firmicutes genome assemblies totaling 470,472, the study selected a custom set of 828 genomes representing major lineages for deeper analysis, Nature reported. Protein similarity network analyses using EGN divided the largest network of 1,442,176 proteins into functional groups using a multilevel community detection algorithm.

For Bacillus species, researchers analyzed all available GenBank genome assemblies numbering 7,377. To manage overrepresented pathogens from the B. cereus and B. anthracis lineages, the team applied Treemmer to prune redundant tree leaves.

This pruning yielded a non-redundant set of 4,774 Bacillus assemblies encoding at least 75 of 84 targeted single-copy orthologues. These assemblies were subsequently used for species tree inference under the -lg -gamma model in FastTree.

Cell surface anchoring domains often display broader phylogenetic distribution and share higher sequence conservation than the functional domains of anchored proteins. This structural conservation can cause network hairballs due to frequent anchoring domain swapping, masking distinct evolutionary histories.

To resolve this, researchers removed common anchoring domain regions like CWB2, SLH, and SH3 from full-length protein sequences using the Bio.Seq module in Biopython. The resulting anchorless functional domain sequences were analyzed separately to map true homology networks.

Structural predictions using AlphaFold2 in Colab Notebook verified the correct removal of these adjacent domain repeats and linkers. Protein networks were generated using BLAST and optimized in EGN with specific thresholds for hit identity and coverage.

About the author: Business Desk

Business Desk at Archynewsy.