How to Submit WGS Genomes

Introduction

DDBJ/EMBL/GenBank accepts both complete and incomplete genomes. Whole Genome Shotgun (WGS) sequencing projects are incomplete genomes or incomplete chromosomes that are being sequenced by a whole genome shotgun strategy. See the WGS project info section for details on what is and is not suitable for submission as a WGS project.

The pieces of a WGS project are the contigs (overlapping reads), and they do not include any gaps. An AGP file can be submitted to indicate how the contig sequences are assembled together into scaffolds (contig sequences separated by gaps) and/or chromosomes. We must have the contig sequences without gaps as the basic units for all WGS projects.

WGS projects may be annotated, but annotation is not required. NCBI has a publicly available Prokaryotic Genomes Annotation Pipeline (PGAAP) . This pipeline generates files that are ready for submission to GenBank, although the submitter is welcome to edit them before submission to GenBank. Submit to PGAAP first, and then submit to GenBank. See below for details about PGAAP submissions to GenBank.

Information about the requirements for more complex assemblies, such as those with PARs or alternate loci, is in the Assembly Submission pages.

WGS genomes without annotation, or with PGAAP annotation, require at least two weeks to be processed. Genomes with annotation require at least one month for processing. Please submit your project with enough lead time.

Below are detailed instructions for preparing a simple WGS submission.

Table of Contents

Requirements

1. Register the genome sequencing project with the BioProject DB, if you have not already done this.  Do not register a duplicate BioProject for the same genome.
2. Make the genome assembly data files .
3. Run the commandline program tbl2asn to generate the .sqn file for submission and the validation and discrepancy report files.
4. Fix problems that are indicated in the .val and discrep files. Failure to do this will cause serious delays in processing.
5. If you have higher-level assembly information, scaffolds and/or chromosomes, then generate an AGP file to build those objects from the wgs-contigs.
6. Generate Assembly information files if your assembly contains complete chromosomes, plasmids or linkage groups (including the mitochondrion) or it assembles them from wgs contigs.
7. Submit via the new Genomes (WGS) in the submission portal
8. Submit the unassembled sequence reads to SRA, the Sequence Read Archive. For questions about submitting, contact the SRA HelpDesk via the "Write to the Help Desk" link at the bottom of the SRA page. Be sure to include in your submissions and correspondence the BioProject ID that was assigned when you registered your project with the BioProject database.

Submission Details:

Data files:

The specifics of the file formats are presented in the tbl2asn page.

• generate a template file with submitter and publication information.
• put the contig sequences into fasta format of the sequences. These files have the suffix .fsa. Each sequence has a definition line beginning with a '>' and a unique identifier, eg contig001, contig002, etc. Note that this unique identifier appears in the DEFINITION line in the flatfile view of the record. Please use concise names that do not include length or coverage information. See some example files.
  • These sequences are the contigs made from overlapping reads. The only allowed Ns are internal Ns that represent ambiguous bases. Remove any Ns that are at the end of a contig, and split a sequence into separate contigs at any Ns that represent gaps. Ns that represent gaps are not allowed, so sequences containing runs of Ns that represent gaps need to be split into individual contigs. This is even true for sequences that were assembled by assemblers that connect mate-pairs. That higher-level assembly information can be presented in an AGP file . The .fsa files can have up to 10,000 sequences per file. Larger submissions need to be split into multiple files.
  • Submit only contigs >199nt.
  • The source information can be included in the defline of each contig or in the tbl2asn commandline. It is included in the same format, in either place. At a minimum the source information is the organism and the relevant strain, breed, cultivar or isolate, if one exists for the sequenced organism. Note that the organism for microbial genomes (prokaryotes and fungi) still includes the strain, eg Escherichia coli BCE032-DM-B or Saccharomyces cerevisiae FL100, so the strain name appears in the organism and also in the strain, as seen in the example below. Bacterial (and other) isolates should also include the isolation source, eg the host or environmental isolation_source, when known. Bacterial annotated assemblies should also include the genetic code, [gcode=11], to reduce the number of false errors.

    Human pathogens should include:

    • host, eg Homo sapiens or spinach or Bos taurus
    • isolation source. This should be non-identifiable information, eg "stool sample from patient with fever" or "blood"
    • collection_date
    • country where the sample was isolated
    • latitude/longitude, in decimal coordinate system, where the sample was collected
    Here is an example of the source information for an annotated bacterial submission:

    [organism=Clostridium difficile ABDC] [strain=ABDC] [host=Homo sapiens] [country=Canada: Montreal] [collection_date=2008] [isolation-source=stool sample] [note=isolated from an outbreak in Montreal] [gcode=11]

  • Contigs that are part of a plasmid, or an organellar chromosome, or specific nuclear chromosomes need to have that information included in the fasta definition line, in these formats:
    • [plasmid-name=pBR322]
    • [plasmid-name=unnamed] (when the plasmid name is not known. However, be sure that each plasmid has a unique name. )
    • [location=mitochondrion]
    • [location=chloroplast]
    • [chromosome=2] (this is usually included only for smaller genomes when the chromosome assignment is certain; don't do this for large, eg mammalian, genomes when reassembly may result in a contig's assignment to a different chromosome)
• quality scores of the sequences. These files correspond to and have the same basenames as the .fsa files, but have the suffix .qvl. The quality scores are optional, but desired.
• annotation files, if appropriate. These correspond to and have the same basenames as the .fsa files, but have the suffix .tbl. The .tbl files have a 5-column tab-delimited format, as described in the annotation instruction pages.

  Be sure to read the annotation requirements in the appropriate annotation guidelines:

  • Prokaryotic Genome Guidelines
  • Eukaryotic Genome Guidelines
  .

  You may be interested to know that NCBI has a publicly available Prokaryotic Genomes Annotation Pipeline (PGAAP). This pipeline generates files that are ready for submission to GenBank, although the submitter is welcome to edit them before submission to GenBank. Submit to PGAAP first, and then submit to GenBank. See below for details about PGAAP submissions to GenBank.

Run tbl2asn:

Put all the files in the same directory, and run tbl2asn (version 19.6 or higher). The basic commandline is:

• tbl2asn -p path_to_files -t template -M n -Z discrep

• if you put the template file in a different directory, then include the full path to that file
• if you want to use a particular file, then use -i file_name, instead of -p path_to_files
• including "-Z discrep" runs the discrepancy report , which looks for inconsistencies or other suspicious problems, so is most appropriate for annotated submissions.
• You can include the source information in the definition line of each contig, as described in the source info above. Alternatively, all of the common source information, can be included with -j in the tbl2asn commandline. Note that the organism for microbial genomes (prokaryotes and fungi) still includes the strain, eg Escherichia coli BCE032-DM-B or Saccharomyces cerevisiae FL100, so the strain name appears in the organism and also in the strain, as seen in the example below. In addition, if the submission is an annnotated prokaryotic genome, then include the genetic code with -j in the commandline:
  • tbl2asn -p path_to_files -t template -M n -Z discrep -j "[organism=Clostridium difficile ABDC] [strain=ABDC] [host=Homo sapiens] [country=Canada: Montreal] [collection_date=2008] [isolation-source=stool sample] [note=isolated from an outbreak in Montreal] [gcode=11]"

Check the Output of the Validation and Discrepancy Report and Fix Problems

• Check the errorsummary.val file for the number, severity and type of errors that are present in the .val files. All Errors and Rejects need to be fixed (as of March 2012, and v19.6 or higher). The presence of errors will slow processing.  Contact genomes@ncbi.nlm.nih.gov with any questions about the validation output.  During processing there may be some questions about other aspects of the submission.
• Check the file named 'discrep' for the results of the discrepancy report. Categories prefaced with FATAL are always unacceptable and must be fixed (FATAL tags were added in January 2012 ).  Some of the categories are informational.  Reports that are not flagged as fatal need to be evaluated to determine if they represent annotation artifacts that need to be corrected or if they are acceptable due to the biology of the genome. See the discrepancy report examples and explanations for guidance. Write to genomes@ncbi.nlm.nih.gov and send the discrep file with questions about this report.
• Make any necessary fixes to the input .fsa and/or .tbl files and run tbl2asn again. Or make the necessary fixes directly to the .sqn file by opening it in Sequin and editing the features there.

AGP file

AGP files provide the ordering and orientation information to construct supercontigs or scaffolds from contigs, or to construct chromosomes from scaffolds and/or contigs.  The AGP file defines these genome assemblies, so include all wgs-contigs that are considered to be part of the genome in the AGP file.

see details

Plan for adopting v2.0 of the AGP specifications

NCBI is switching from version 1.1 of the AGP specification to the new version, version 2.0, because the latter can convey valuable information on the nature of the evidence linking sequences on either side of a gap. Files in AGP2.0 will be accepted beginning Feb. 10, 2012 and will be required beginning July 1, 2012. See the full announcement.

Assembly Information Files

Additional files are needed if the genome assembly includes chromosomes (or plasmids or linkage-groups) or unlocalized scaffolds (scaffolds known to be part of a particular chromosome/plasmid/linkage-group, but whose location is not known).

see details

Submitting PGAAP-annotated genomes

You submit to PGAAP first, before you prepare your GenBank WGS submission. Note that PGAAP submissions require a different file format than WGS submissions. See the detailed PGAAP instructions here, http://www.ncbi.nlm.nih.gov/genomes/static/Annotation_pipeline_README.txt

PGAAP will provide you with the annotated contig files which will be in the correct format for a WGS submission. You can review the annotation and make any changes that you think are necessary. If you do not make any changes, then you just need to send a request to genomes@ncbi.nlm.nih.gov to submit your PGAAP-annotated genome to GenBank, and include the directory name for the PGAAP-generated files. We will get the files from there and process them.

If you do make changes to the annotation, then you will need to generate a .sqn file as described in the Run tbl2asn section, and then submit via the new Genomes (WGS) in the submission portal. Be sure to include in the comment box of the submission that the annotation is from PGAAP with your modifications.

NOTE: In correspondence about using the PGAAP resource, include "PGAAP" in the subject line to be sure your messages are seen by the correct group.