STBase: one million species trees for comparative biology

Area of Science:

• Evolutionary Biology
• Bioinformatics
• Genomics

Background:

• Phylogenetic trees are crucial for comparative evolutionary biology, but mismatches exist between available data and published phylogenies.
• Gene trees often require adaptation for species-level comparisons due to discordance and gene duplication.

Purpose of the Study:

• To develop a database (STBase) for rapid retrieval of species-level phylogenetic hypotheses.
• To facilitate comparative biologists' access to relevant phylogenetic data for their taxa of interest.

Main Methods:

• Assembled 1 million single- and multi-locus datasets with 1000 putative species trees from GenBank data.
• Leveraged theoretical work to prune gene trees and construct conflict-free species trees.
• Ameliorated missing data impacts by assembling only decisive datasets.

Main Results:

• STBase allows quick retrieval of species-level phylogenetic hypotheses based on user queries.
• Retrieval times are consistently fast, typically a few seconds, regardless of database size.
• Tree quality is assessed in real-time using bootstrap support on overlapping subtrees.

Conclusions:

• STBase provides a valuable tool for comparative biologists to access relevant phylogenetic data.
• The database serves as a prototype for future species tree-focused databases.
• It aids in the assembly of larger species phylogenies from precomputed trees.