Sea doggo go blub? Placing new prokaryotic lineages on the tree of life

In about two decades since metagenomics and single cell genomics became common practice in microbial ecology, a deluge of genomes from uncultivated taxa with immense taxonomic and metabolic novelty has become available. The genomic data also brought about taxonomic classification approaches that use phylogenomics with concatenations of tens to hundreds of marker proteins. However, with deeper phylogenies and more divergent lineages included in the tree of life comes a host of phylogenetic systematic errors (e.g., long branch attraction, compositional bias). Most microbiologists lack the experience to perform the advanced phylogenomic analyses required to address these errors and instead rely on automated tools for the taxonomic assignment of their new lineages. This in turn creates issues, if these phylogenies are used to formulate evolutionary hypotheses.

To address these issues, we have created WhereDoGGo (Where Does my Genome Go?), a phylogenomics pipeline that aims to address different sources of error in the placement of prokaryotic lineages on the tree of life, while being user-friendly enough to be used by researchers with minimal bioinformatics experience.

WhereDoGGo leverages the normalized prokaryotic taxonomy system of the Genome Taxonomy Database to pick genomes with a balanced and diverse taxonomic sampling. It can create concatenated alignments of several published market sets and employs IQ-TREE as a framework to perform a series of analyses aiming to address the different sources of systematic error in phylogenies.

In this talk, we will first present the structure and explain the use of WhereDoGGo. Then, we will demonstrate its utility through a series of test cases, including the phylogenomic placement of novel lineages from coastal hot springs in the South Aegean Volcanic Arc.