Starship giant transposable elements cluster by host taxonomy using kmer-based phylogenetics

Abstract

Starships are a recently established superfamily of giant cargo-mobilising transposable elements in the fungal subphylum Pezizomyotina (phylum Ascomycota). To date, Starship elements have been identified up to ∼700 Kbp in length and carrying hundreds of accessory genes, which can confer both beneficial and deleterious traits to the host genome. Classification of Starship elements is centred on the tyrosine recombinase gene that mobilises the element, termed the captain. We contribute a new perspective to Starship relatedness by using an alignment-free kmer-based phylogenetic tree building method, which can infer relationships between elements in their entirety, including both active and degraded elements and irrespective of high variability in element length and cargo content. In doing so we found that relationships between entire Starships differed from those inferred from captain genes and revealed patterns of element relatedness corresponding to host taxonomy. Using Starships from root/soil-dwelling Gaeumannomyces species as a case study, we found that kmer-based relationships correspond with similarity of cargo gene content. Our results provide insights into the prevalence of Starship-mediated horizontal transfer events. This novel application of a kmer-based phylogenetics approach overcomes the issue of how to represent and compare highly variable Starship elements as a whole, and in effect shifts the perspective from a captain to a cargo-centred concept of Starship identity.

Graphical Abstract

cargo-mobilising elements, Gaeumannomyces, Ascomycota, Pezizomycotina, Fungi

Accepted manuscripts

Accepted manuscripts are PDF versions of the author’s final manuscript, as accepted for publication by the journal but prior to copyediting or typesetting. They can be cited using the author(s), article title, journal title, year of online publication, and DOI. They will be replaced by the final typeset articles, which may therefore contain changes. The DOI will remain the same throughout.

Author notes

Present address: Daniel Smith, Department of Computational and Systems Biology, John Innes Centre, Norwich, Norfolk, NR4 7UH, UK