DDBJ | Contents

What is DDBJ

DDBJ is the annotated nucleotide sequence database that forms part of INSDC (DDBJ / NCBI GenBank / EBI ENA). Registered entries are converted to flat-file format and exchanged daily among the three INSDC members.

Two submission windows feed into the same database:

Window	Purpose
NSSS (DDBJ Nucleotide Sequence Submission System)	Web wizard for small / standard entries
MSS (Mass Submission System)	File-based upload (FASTA + annotation TSV) for large-scale, full-replicon, or BioProject/BioSample-linked data

Raw sequencer reads do not belong here; submit them to DRA instead.

NOTE

If you are unsure which window to use, the Submit Navigator walks you through your data type and recommends NSSS or MSS.

Data accepted

Submissions are sorted by INSDC division (HUM / PRI / ROD / MAM / VRT / INV / PLN / BCT / VRL / PHG / ENV / SYN) and data category. The MSS application form offers the following data types:

Data type	Description
WGS	Whole Genome Shotgun (draft genome)
GNM	Finished level genome sequence, non-WGS
MAG	Metagenome-Assembled Genome
SAG	Single Amplified Genome
TLS	Targeted Locus Study (16S rRNA and other targeted loci, bulk)
HTG	High Throughput Genomic Sequences
TSA	Transcriptome Shotgun Assembly
HTC	High Throughput cDNA Sequences
EST	Expressed Sequence Tags
GSS	Genome Survey Sequences
TPA	Third Party Data
MISC	Anything else (standalone organelle / plasmid / virus / phage genome, etc.)

NSSS only handles "standard entries" (cDNA, genome fragments, single 16S rRNA, vector sequences, etc.) below replicon scale.

MSS

Mass Submission System (MSS) accepts comparatively large-scale nucleotide data via file upload. Use MSS whenever NSSS cannot cover the case:

Special data categories: EST / TSA / HTC / GSS / HTG / WGS / TLS / TPA
Scale: any of a single sequence >= 500 kb, >= 30 Features in a single sequence, or more than 100 sequences in total
Full-length replicons (finished or draft): genomes, chromosomes, organelle genomes, virus / phage genomes / segments, plasmids
BioProject / BioSample linkage required: metagenomes (MAG / SAG), environmental profiles, sequences derived from the same strain as a planned or registered full-replicon genome, prokaryotic 16S rRNA reports, etc.

The application form lives at https://mss.ddbj.nig.ac.jp/. Upload via the browser, via SFTP, or by specifying a DFAST job ID. Use SCP / SFTP when the total upload exceeds 10 GB.

MSS file formats

Sequence file: FASTA-like text. Header lines start with >, followed by the nucleotide sequence; entries are delimited by //. Entry names must be unique ASCII alphanumerics within 32 characters, and must not contain space, ", =, |, >, [], or \. Strip trailing n characters. Allowed extensions: .fasta / .seq.fa / .fa / .fna / .seq.
Annotation file: tab-separated, 5 columns (Entry / Feature / Location / Qualifier / Value). Allowed extensions: .ann / .annt.tsv / .ann.txt. Provide the COMMON section (SUBMITTER / REFERENCE / DATE / DBLINK, etc.) together with each entry's Biological Features (source / CDS / rRNA / tRNA / ncRNA, etc.). Do not prefix qualifiers with /.
AGP file: CON entries only. New CON submissions are no longer accepted.

WARNING

File names must not contain multibyte characters, spaces, backticks, <>, or (). Email attachments are discouraged; always upload through the MSS form.

NSSS

NSSS (DDBJ Nucleotide Sequence Submission System) is the browser-based system for annotated nucleotide sequence registration. It is the recommended window for small / standard entries that fall outside MSS, and the recommended starting point for first-time submitters.

Entry page: https://www.ddbj.nig.ac.jp/ddbj/web-submission.html
Start a new submission: https://ddbj.nig.ac.jp/submission
Help: https://www.ddbj.nig.ac.jp/ddbj/web-submission-help.html
Supported browsers: Google Chrome / Microsoft Edge / Mozilla Firefox

NSSS handles new submissions only; corrections to released accessions go through the DDBJ update request form. TPA-Exp and TPA-Inf have been retired from NSSS since January 2025, and TPA:assembly is handled exclusively by MSS.

Bookmark the page after clicking Next on each step; the bookmark resumes the submission later.

When to use MSS vs NSSS

NSSS and MSS are two routes into INSDC and produce the same INSDC flat files in the end.

Axis	Prefer NSSS	Prefer MSS
Category	Standard entries (cDNA, genome fragments, a single 16S rRNA, vector sequences, etc.)	EST / TSA / HTC / GSS / HTG / WGS / TLS / TPA
Scale	< 500 kb per sequence, < 30 Features, < 100 sequences total	A sequence >= 500 kb, >= 30 Features, or more than 100 sequences total
Completeness	Single or partial sequences	Replicon-scale (complete genomes, chromosomes, organelles, plasmids, virus / phage genomes or segments)
Linkage	BioProject / BioSample linkage optional	BioProject / BioSample linkage required (metagenomes, TPA, same-strain cross-references, prokaryotic 16S rRNA reports, etc.)

When in doubt, default to NSSS and only switch to MSS when one of the conditions above applies.

Submission flow

MSS

Create a DDBJ account / sign in via D-way.
Obtain a BioProject ID, BioSample ID, and locus_tag prefix in advance where applicable.
Prepare the FASTA and annotation TSV files (for prokaryotic genomes, DFAST can generate them automatically).
Validate the format and CDS translations locally with UME / Parser / transChecker.
Sign in to the MSS form and upload the files; a Mass-ID (for example [DDBJ:NSUB000001]) is issued and notified by email.
A DDBJ curator reviews the submission against INSDC rules and contacts you by email for revisions or confirmation.
Once finalised, the accession numbers are sent to the contact person by email.
The entries are released on the scheduled release date (immediately or after a hold) and distributed across INSDC.

NSSS

Create a DDBJ account / sign in via D-way.
Remove vector / adapter contamination, for example with VecScreen.
Start from https://ddbj.nig.ac.jp/submission by clicking Create new submission.
Fill in submitter, REFERENCE, sequence, and Feature / Qualifier details in the web form (the Submission ID lets you pause and resume).
A curator reviews the submission, emails the accession numbers, and releases the entries on the scheduled date.

Prerequisites

A DDBJ account (D-way). SCP / SFTP transfer additionally requires a registered public key.
Contact person information (name, affiliation, address, phone). Multiple contacts (at minimum the operator and a supervisor) are strongly recommended so future correspondence does not stall.
For genome-scale, metagenome, TPA, and other linkage-required cases, obtain a BioProject ID and BioSample ID beforehand.
If the annotation uses locus_tag, reserve a locus_tag prefix when registering the BioSample (prefixes cannot be changed after issuance).
Decide the scheduled release date (immediate release or hold).
MSS-only local validation tools:
- UME (format + CDS translation; Windows / Linux / macOS)
- Parser (format; Linux)
- transChecker (CDS translation; Linux)

IMPORTANT

A single MSS Submission may only bundle data sharing the same contact person, data category, and scheduled release date. Split data that differ on any of these into separate Submissions.

Accession numbers

Category	Format	Example
Single entry (legacy)	1 letter + 5 digits	`A12345`
Single entry (conventional)	2 letters + 6 digits	`AB123456`
Single entry (extended)	2 letters + 8 digits	`AB12345678`
WGS / TSA / TLS	4 letters + 8-10 digits	`BAAA01000000` (master) / `BAAA01000001`- (contigs)
WGS / TSA / TLS (new format)	6 letters + 9-11 digits	`ABCDEF010123456`
MGA	5 letters + 7+ digits	`ABCDE1234567`
protein_id	3 letters + 5 or 7 digits	`ABC12345` / `ABC1234567`
BioProject	`PRJDB` + digits	`PRJDB12345`
BioSample	`SAMD` + 8 digits	`SAMD00000001`

During submission the working identifier is the Mass-ID (for example [DDBJ:NSUB000001]) for MSS or the Submission ID for NSSS. Final accession numbers are emailed to the contact person once curation completes.

Metagenome (MAG / SAG)

Metagenome-derived data are registered as genome entries in the ENV division. BioProject and BioSample must be registered in advance, the INSDC-prescribed qualifiers must be present, and the submission window is MSS.

MAG (Metagenome-Assembled Genome)

Use the BioSample MIMAG package and an organism name (without uncultured) representative of the MAG. Record the source metagenome via the metagenome_source attribute (for example soil metagenome) and the derived_from reference to the metagenome sample.
The source feature must carry /metagenome_source, /environmental_sample, /isolation_source, /isolate, /organism, and /mol_type="genomic DNA".
/strain must not be used. Host information goes in /host.
Organism names containing metagenome or uncultured are not allowed in /organism.
The ST_COMMENT block must include Assembly Method, Genome Coverage, and Sequencing Technology (plus Assembly Name for eukaryotes).

SAG (Single Amplified Genome)

Use the BioSample MISAG package.
The source feature must carry /note="single amplified genome" and /isolation_source.
For co-assemblies of multiple cells, state this in /note and reference the source samples via the BioSample derived_from attribute.

Haplotype assembly

When submitting a diploid genome as separate Principal and Alternate haplotypes, use the following three-layer structure.

Layer	Role
Principal BioProject	Submission unit for the principal haplotype
Alternate BioProject	Submission unit for the alternate haplotype
Umbrella BioProject	Parent project that ties the two above together

Record Diploid :: Principal haplotype or Diploid :: Alternate haplotype in ST_COMMENT.
Both haplotypes must reference the same BioSample (use the MIGS package).
Submitters choose the labelling convention: Principal / Alternate, Haplotype 1 / 2, or Maternal / Paternal.
An Umbrella BioProject can also bundle the corresponding raw-reads project under DRA.

TPA (Third Party Annotation)

TPA covers data assembled or (re)annotated by a third party from primary entries that someone else has already deposited. The only submission window is MSS; NSSS does not handle TPA.

WARNING

Since January 2025, DDBJ only accepts TPA:assembly. New submissions for TPA:experimental, TPA:inferential, and TPA:specialist_db are suspended.

Key constraints:

The primary entry must be deposited in INSDC (it may be unreleased at TPA submission time but must be retrievable when the TPA entry is released).
Generation of the assembled sequence must be subject to peer-reviewed journal peer review.
TPA and primary sequences may differ by no more than 5 %.
Gaps without a citation must be 50 bp or shorter.
Full-replicon assemblies (TPA-WGS, etc.) require BioProject and BioSample registration.

BioProject - Project registration (Umbrella BioProject)
BioSample - Sample metadata registration
DRA - NGS raw reads submission