Oral Paper

         Systematics

Using PacBio long-read sequences and the PURC II pipeline to infer a multi-locus reticulate phylogeny of Cystopteridaceae (Polypodiales)

Presenting Author
Chinedum Anajemba
Description
Allopolyploids present a challenge for multilocus phylogenetic inference due to their multiple distinct subgenomes, each of which may have its own evolutionary history. Generating sequence data from such lineages is challenging due to the difficulty in isolating the sequences of each of the distinct homoeologous gene copies. In addition, the phasing of gene copies into polyploid subgenomes remain another challenge even when the correct biological sequences are inferred. Recent advances in polyploid phylogenetics have made it possible to effectively undertake phylogenetic study of groups that comprise polyploids. In this study, we investigated the phylogeny of Cystopteridaceae, a fern family that contains a wide array of polyploid complexes. We used PURC II to infer the underlying sequences of four c. 1-kb-long nuclear loci from a broad sample of Cystopteridaceae accessions generated using the PacBio platform. PURC II (Pipeline for Untangling Reticulate Complexes) infers the true biological sequences (alleles, paralogs, or homeologs) from amplicon data, by using a model-based method (amplicon sequence variant inference) to correct PCR and sequencing errors. The sequences of each polyploid accession recovered by PURC II were assigned to their respective subgenomes using homologizer, a Bayesian method that uses a phylogenetic framework to phase gene copies across loci, allowing us to infer a multilocus reticulate phylogeny of the family.