Using GFF for Genome Annotation Submissions to GenBank

GenBank submission improvements in the last few years have made it easier for users to deposit whole genome sequence data in the public INSDC archives, which provides data standardization and greatly expands access to these important datasets. Whole genome sequence data can be more useful when it incorporates at least basic annotation information about the location of genes and proteins. The resulting annotations are available in many NCBI resources, including the nuccore, protein and BLAST NR databases, and are provided on NCBI's redesigned genomes FTP site in standardized formats, making it advantageous for submitters to provide valid annotation along with their genome sequencing data.  Traditionally, annotation submission to GenBank has used a five-column feature table that is not a typical output of most annotation software. NCBI is developing tools to read annotation from GFF3 or GTF-formatted files, convert it to NCBI's standard formats, and validate the data before submission to GenBank. This talk will discuss basic requirements for using GFF3 or GTF to submit annotation, including an overview of required and optional attributes, and describe markup that will be automatically added if not available in the original submission files.