P1267
Annotating the Nuclear Genome of Azolla filiculoides

Mathew Simenc , California State University Fullerton, Fullerton, CA
Fay-Wei Li , University of California Berkeley, Berkeley, CA
Nils Koppers , Heinrich Heine Universität Düsseldorf, Düsseldorf, Germany
Shifeng Cheng , Beijing Genomics Institute-Shenzhen, Shenzhen, China
Bo Song , Beijing Genomics Institute-Shenzhen, Shenzhen, China
Xin Liu , Beijing Genomics Institute-Shenzhen, Shenzhen, China
Yue Song , Beijing Genomics Institute-Shenzhen, Shenzhen, China
Shanyun Xu , Beijing Genomics Institute-Shenzhen, Shenzhen, China
Laura Dijkhuizen , Utrecht University, Utrecht, Netherlands
Bruno Huettel , Max Planck Genome Centre, Cologne, Germany
Paul Brouwer , Utrecht University, Utrecht, Netherlands
Andrea Bräutigam , Heinrich Heine Universität Düsseldorf, Düsseldorf, Germany
Gane Ka-Shu Wong , Department of Biological Sciences, University of Alberta, Edmonton, AB, AB, Canada
Henriette Schluepmann , Utrecht University, Utrecht, Netherlands
Kathleen M. Pryer , Duke University, Durham, NC
Joshua P. Der , California State University, Fullerton, Fullerton, CA
Ferns are one of the few remaining embryophyte lineages lacking a complete nuclear genome sequence. To remedy this gap, we have sequenced and annotated the nuclear genome of the heterosporous water fern, Azolla filiculoides. Azolla is a fast growing plant that forms an obligate symbiotic relationship with nitrogen-fixing cyanobacteria and has huge potential for carbon sequestration, providing nitrogen-rich feedstock, and as a sustainable biofertilizer. Azolla will serve as a critical lineage in comparative genomic analyses of land plants; in addition, knowledge of its genome content and structure will further our understanding of the evolution of symbiosis. A whole genome shotgun assembly was constructed using 51X PacBio data and was validated using over 200X coverage with Illumina sequence data. Here we present and evaluate an initial draft annotation of Azolla constructed using the MAKER-P pipeline. A custom repeat library was generated using RepeatModeler and LTRHarvest, and repetitive regions of the genome were masked using RepeatMasker prior to gene annotation. Initial gene models were predicted using both SNAP and Augustus and are supported by extensive RNA-seq transcriptome data. A final set of high-confidence gene models was identified using EVidenceModeler. Gene models are assigned putative functions using Gene Ontology and classified into other land plant gene families using OrthoFinder.