How to Submit a Bacterial Genome to NCBI and Write a Genome Announcement

The records you will create

BioProject organises project data submitted to NCBI, EBI and DDBJ and ties related data together across these archives. BioSample captures descriptive information about each biological sample, and both are created through NCBI's web-based Submission Portal ^[1]. The metadata fields follow the Minimum Information about a Genome Sequence (MIGS) specification from the Genomic Standards Consortium ^[2].

Step by step

1. Collect your metadata: strain name, isolation source, host, collection date, geographic origin, and sequencing platform and kit.
2. In the NCBI Submission Portal, register a BioProject and a BioSample for each strain.
3. Upload the raw reads to SRA and link them to the BioSample.
4. Check the assembly before submitting. Remove very short contigs (NCBI does not accept sequences under 200 bp), and screen for adapters and contamination.
5. Submit the assembly to GenBank, either with your own annotation or with a request for NCBI to annotate it with PGAP.
6. Respond to any questions from NCBI staff. Your submission may be flagged for contaminated or adapter sequences that need trimming.
7. Cite the BioProject, SRA and GenBank accession numbers in your manuscript. You can ask for the data to be released when the paper is published.

Annotation with PGAP

PGAP combines alignment-based evidence with gene prediction from the sequence itself. Its gene finder, GeneMarkS+, uses the positions of known proteins and RNAs as a starting map, then adjusts its predictions across the genome. PGAP relies more on similarity where good comparative data exist, and more on statistical prediction where they don't ^[3]. Asking NCBI to annotate your genome keeps it consistent with other prokaryotic genomes in GenBank. You can still use your own annotation, for example from Prokka, in the paper.

Writing a genome announcement

A genome announcement is a short paper that reports a new genome sequence and its accession numbers. Journals such as Microbiology Resource Announcements publish them. Editors and reviewers mostly check that the methods are complete and that the data are public, or will be released on publication. A typical announcement contains:

• where and how the strain was isolated, and why it is of interest
• DNA extraction, library preparation and sequencing platform
• every software tool used for QC, assembly and annotation, with versions and non-default parameters
• genome statistics: size, number of contigs, N50, GC content, coverage, and counts of coding sequences, rRNAs and tRNAs
• the BioProject, BioSample, SRA and GenBank accession numbers

Before submitting, check the quality of the assembly using the metrics in assembly quality metrics, and confirm the species name with ANI. A complete genome from long-read or hybrid sequencing makes a stronger announcement than a fragmented draft.

How BacSeq Analytics helps

BacSeq Analytics reports include the genome statistics and methods details that a genome announcement needs, and the Premium package includes a consultation for questions about your submission or manuscript. See the packages for what each level includes.