Submitting sequences to dbGSS and GenBank
The GSS division of GenBank is similar in nature to the EST division, except that its sequences are genomic in origin, rather than cDNA (mRNA). The GSS division contains (but is not limited to) the following types of data:
- random "single pass read" genome survey sequences.
- single pass reads from cosmid/BAC/YAC ends (these could be chromosome specific, but need not be)
- exon trapped genomic sequences
- Alu PCR sequences
Section 1.3.3 of the GenBank 96.0 release notes provides additional information about the GSS division.
GSSs by nature are usually submitted to GenBank and dbGSS as batches of dozens to thousands of entries, with a great deal of redundancy in the citation, submittor and library information. To improve the efficiency of the submission process for this type of data, we have designed a separate streamlined submission process and data format.
NOTE: Beginning in 2009 Sequences derived from "next generation" sequencing platforms, including Roche 454, Illumina, Applied Biosystems SOLiD, and Helicos Biosciences HeliScope, should be submitted to the Short Read Archive (SRA) (For information contact sra@ncbi.nlm.nih.gov.)
Data submission file types
There are two parts to the submission instructions, one for the sequence data, and one for any mapping data. (NOTE: starting in 2009 map data will no longer be entered for dbGSS submissions.)
The batch submission process for GSS sequence data involves the completion of four file types:
1. Publication
2. Library
3. Contact
4. GSS sequence file
The format for each file is described below.
If all the GSSs share the same Publication, Library, and Contact information, you only need to prepare one of each of those files. Then complete a separate GSS file (file type d) for each sequence.
If any of the GSS files have different Publication, Library, or Contact information, you must complete a new set of file types 1-3.
Once we have entered particular Publication, Library, or Contact information into the database, you do not need to resend the data input files.
Mailing files to dbGSS
Send the completed files to: batch-sub@ncbi.nlm.nih.gov
You can attach all the files to a single email message, or you can include them in the body of the email message. Please be sure that they are in plain text (ASCII) format.
We prefer to have the individual GSS batched together as much as possible.
You can submit Publication, Library, and Contact data together in one file. You can also send them in the same file as the GSS entries - the TYPE field will differentiate them for the parsing software.
Assignment of GenBank Accession numbers and release of data
You will receive a list of dbGSS IDs and GenBank Accession numbers from a dbGSS curator via email.
If you would like your sequences held confidential until publication, you can indicate that by putting the release date in the PUBLIC field of the GSS files. Your sequences will be released on that date, or when the Accession numbers or sequence data are published, whichever comes first.
Once your sequences are released into the public database, they will be available from the GSS division of GenBank (accessible through the Entrez Nucleotide division).
Updating your dbGSS data
Updates to GSS entries are done basically in the same way as new entries. Changes to any item in the GSS input file (other than GSS# or CONT_NAME) are made by completing an input file with new data in the fields that need to be changed. For the STATUS field, enter "Update" instead of "New".
In addition to the fields to be changed Updates need to include TYPE, STATUS, GSS#, and CONT_NAME fields.
For changes in Publication, Contact, or Source data, or for changes in GSS#'s or CONT_NAME, send an email message describing the change that is needed.
Send the update files to: batch-sub@ncbi.nlm.nih.gov
Questions and Comments
If you have questions about the GSS submission format, please contact info@ncbi.nlm.nih.gov
Submission Format for GSS Sequence Data
The following is a specification for flat file formats for delivering GSS and related data to the NCBI GSS database.
- The format consists of colon delineated capitalized field tags, followed by data.
- The data fields should appear on the same line as the tag, with no line wrapping. Exceptions to this are the TITLE and AUTHORS fields of the Publication file; Description (DESCR:) field of the Library file; and the CITATION, COMMENT, and SEQUENCE fields of the GSS file. In these fields, the data text begins on the same line as the tag or on the line following the field tag, and the text can cover several lines.
- Note that some fields are obligatory.
- The TYPE field is obligatory at the beginning of each entry, even if there are multiple entries of a given type in a file.
- If data are not available for a non-obligatory field, the field can either be omitted entirely, or the tag may be included with an empty data field. Please do not put "*", "-", etc. to indicate missing data.
- Each record (including the last record in the file) should end with a double-bar tag (||) to indicate the end of the record.
- Please also see the bulleted tips at the end of each file format for important notes about that file type.
File Types
There are four types of deliverable files:
1. Publication
2. Library
3. Contact
4. GSS sequence file
Each GSS file needs to reference the Publication, Library, and Contact data. Therefore the Publication, Library, and Contact files must be in the database when the GSS file is entered. Once these files have been submitted and entered, they do not need to be re-submitted for additional GSS files that have the same Publication, Library, or Contact.
1. Publication Files
The following is an example of the valid tags and some illustrative data:
TYPE: Entry type - must be "Pub" for publication entries.
**Obligatory field**.
MEDUID: Medline unique identifier.
Not obligatory, include if you know it.
TITLE: Title of article.
**Obligatory field**.
Begin on tag line or on line below tag, use multiple lines if needed
AUTHORS: Author name, format: Name,I.I.; Name2,I.I.; Name3,I.I.
**Obligatory field**.
Begin on tag line or on line below tag, use multiple lines if needed
JOURNAL: Journal name
VOLUME: Volume number
SUPPL: Supplement number
ISSUE: Issue number
I_SUPPL: Issue supplement number
PAGES: Page, format: 123-9
YEAR: Year of publication.
**Obligatory field**.
STATUS: Status field. Use 1, 2, 3, or 4.
1=unpublished, 2=submitted, 3=in press, 4=published
**Obligatory field**.
||
Examples:
TYPE: Pub MEDUID: 92347897 TITLE: Genomic sequences from a subtracted retinal pigment epithelium library AUTHORS: Gieser,L.; Swaroop,A. JOURNAL: Genomics VOLUME: 13 ISSUE: 2 PAGES: 873-6 YEAR: 1992 STATUS: 4 ||
- The TYPE field is obligatory at the beginning of each entry, even if there are multiple entries of a given type in a file.
- The MEDUID field is a MEDLINE record unique identifier. We do not normally expect you to supply this. We try to retrieve this from our relational version of MEDLINE database.
- The STATUS field is 1=unpublished, 2=submitted, 3=in press, 4=published
- The TITLE field is a free format string. The only requirement is that you put an identical string in the CITATION field of the GSS files, because we will be matching that field automatically against the publications in the publication table and replacing the string with the publication identity number in the GSS table.
2. Library Files
The following is an example of the valid tags and some illustrative data:
TYPE: Entry type - must be "Lib" for library entries.
**Obligatory field**.
NAME: Name of library.
**Obligatory field**.
ORGANISM: Organism from which library prepared.
STRAIN: Organism strain
CULTIVAR: Plant cultivar
ISOLATE: Individual isolate from which the sequence was obtained
SEX: Sex of organism (female, male, hermaphrodite)
ORGAN: Organ name
TISSUE: Tissue type
CELL_TYPE: Cell type
CELL_LINE: Name of cell line
STAGE: Developmental stage
HOST: Laboratory host
VECTOR: Name of vector
V_TYPE: Type of vector (Cosmid, Phage, Plasmid, YAC, other)
RE_1: Restriction enzyme at site1 of vector
RE_2: Restriction enzyme at site2 of vector
DESCR: Description of library preparation methods,
vector, etc.
This field starts on the same line or on the line below the DESCR: tag.
||
Examples:
TYPE: Lib NAME: Rat Lambda Zap Express Library ORGANISM: Rattus norvegicus STRAIN: Sprague-Dawley SEX: male STAGE: embryonic day 17 post-fertilization TISSUE: aorta CELL_TYPE: vascular smooth muscle DESCR: Put description here. ||
- The TYPE field is obligatory at the beginning of each entry, even if there are multiple entries of a given type in a file.
- Try to keep the library NAME field to <= 48 characters. We can accept up to 255 characters, but it will be truncated to 48 characters in the identification line of the FASTA file created for BLAST searching.
- When you enter the library NAME in the Library Files, please note that the identical string must be used in the LIBRARY field of the GSS files.
- The DESCR field should contain as much detail about the library as seems appropriate.
3. Contact Files
The following is an example of the valid tags and some illustrative data:
TYPE: Entry type - must be "Cont" for contact entries.
**Obligatory field**.
NAME: Name of person providing the GSS sequence
**Obligatory field**.
FAX: Fax number as string of digits.
TEL: Telephone number as string of digits.
EMAIL: E-mail address
LAB: Laboratory
INST: Institution name
ADDR: Address string
||
Examples:
TYPE: Cont NAME: Sikela JM FAX: 303 270 7097 TEL: 303 270 EMAIL: tjs@tally.hsc.colorado.edu LAB: Department of Pharmacology INST: University of Colorado Health Sciences Center ADDR: Box C236, 4200 E. 9th Ave., Denver, CO 80262-0236, USA ||
- The TYPE field is obligatory at the beginning of each entry, even if there are multiple entries of a given type in a file.
- None of the other fields are obligatory, but we require at least the name of a contact person.
- We would like as many of the fields filled in as possible to provide complete information to the user for contacting a source for the GSS or further information about it.
- The contact name field (CONT_NAME) in the GSS files must contain an identical string to the string used in the NAME field of the Contact file, for automatic matching.
4. GSS Files
The following is an example of the valid tags and some illustrative data:
TYPE: Entry type - must be "GSS" for GSS entries.
**Obligatory field**
STATUS: Status of GSS entry - "New" or "Update".
**Obligatory field**
CONT_NAME: Name of contact
Must be identical string to the contact entry
**Obligatory field**
CITATION: Journal citation
Must be identical string to the publication title
Begins on line below tag.
Use continuation lines if needed.
**Obligatory field**
LIBRARY: Library name
Must be identical string to library name entry.
**Obligatory field**
GSS#: GSS name or number assigned by contact lab. For GSS entry
updates, this is the string we match on.
**Obligatory field**
GDB#: Genome Database accession number
GDB_DSEG: Genome Database Dsegment number
CLONE: Clone number/name
SOURCE: Source providing clone, e.g., ATCC
SOURCE_DNA: Source identity number for the clone as pure DNA
SOURCE_INHOST: Source identity number for the clone stored in the host
OTHER_GSS: Other GSSs on this clone.
DBNAME: Database name for cross-reference to another
database
DBXREF: Database cross-reference accession
PCR_F: Forward PCR primer sequence
PCR_B: Backward PCR primer sequence
INSERT: Insert length (in bases)
ERROR: Estimated error in insert length (bases)
PLATE: Plate number or code
ROW: Row number or letter
COLUMN: Column number or letter
SEQ_PRIMER: Sequencing primer description or sequence
P_END: Which end sequenced, e.g., 5'
HIQUAL_START: Base position of start of high-quality sequence
(default = 1)
HIQUAL_STOP: Base position of last base of high-quality
sequence
DNA_TYPE: Genomic (default), cDNA, Viral, Synthetic, Other
CLASS: Class of sequencing method, e.g., BAC ends,
YAC ends, exon-trapped
**Obligatory field**
PUBLIC: Date of public release
Leave blank for immediate release.
**Obligatory field**
Format: MM/DD/YYYY
PUT_ID: Putative identification of sequence by submitter
COMMENT: Comments about GSS.
Text starts on same line or on the line below COMMENT: tag.
SEQUENCE: Sequence string.
Text starts on same line or on the line below SEQUENCE: tag.
**Obligatory field**
||
Examples:
TYPE: GSS STATUS: New CONT_NAME: Sikela JM GSS#: Ayh00001 CLONE: HHC189 SOURCE: ATCC SOURCE_INHOST: 65128 OTHER_GSS: GSS00093, GSS000101 CITATION: Genomic sequences from Human brain tissue SEQ_PRIMER: M13 Forward P_END: 5' HIQUAL_START: 1 HIQUAL_STOP: 285 DNA_TYPE: Genomic CLASS: shotgun LIBRARY: Hippocampus, Stratagene (cat. #936205) PUBLIC: PUT_ID: Actin, gamma, skeletal COMMENT: SEQUENCE: AATCAGCCTGCAAGCAAAAGATAGGAATATTCACCTACAGTGGGCACCTCCTTAAGAAGCTG ATAGCTTGTTACACAGTAATTAGATTGAAGATAATGGACACGAAACATATTCCGGGATTAAA CATTCTTGTCAAGAAAGGGGGAGAGAAGTCTGTTGTGCAAGTTTCAAAGAAAAAGGGTACCA GCAAAAGTGATAATGATTTGAGGATTTCTGTCTCTAATTGGAGGATGATTCTCATGTAAGGT GCAAAAGTGATAATGATTTGAGGATTTCTGTCTCTAATTGGAGGATGATTCTCATGTAAGGT TGTTAGGAAATGGCAAAGTATTGATGATTGTGTGCTATGTGATTGGTGCTAGATACTTTAAC TGAGTATACGAGTGAAATACTTGAGACTCGTGTCACTT ||
- The TYPE field is obligatory at the beginning of each entry, even if there are multiple entries of a given type in a file.
- Valid data values for the GSS STATUS field are New (new entry) or Update (change existing GSS entry).
- When updating a GSS, only the fields present in the GSS file will be changed.
- The DNA_TYPE is assumed to be Genomic, so this field may be omitted unless the DNA type differs from this.
- Sequences can start on the same line or on the line below the SEQUENCE field tag.
AFLP fragment Alu-PCR B1-PCR BAC ends BAC sequence gap BAC subclone BAC subclone end BAC/YAC ends CAPS CoT 5E-3 hydroxyapatite-fractioned DNA Concatamer T-DNA junction DArT clone Ds tagged Ds/TDNA launch pad ERIC-PCR EcoRI fragments Gene Trap Genomic PCR High-Cot HindIII fragments HpaII fragments HpaII/MspI fragment Hydroxyapatite-fractionated DNA ISSR Intron Spanning Low-Cot MboI fragments MuTAIL-PCR NdeI/DraI fragments NotI site P1 ends PAC end PAC nested deletions PAC subclone PCR fragment PCR from cDNA PCR product PCR product with degenerate primers PCR with nonspecific primers PCR with specific primers PCR-based subtractive hybridization PSTI fragment RAPD REP-PCR RFLP clone RFLP probe RLGS Random amplified microsatellites Random complexity-reduced genomic representation Random sheared small inserts SCAR SRAP SSR-containing BAC subclone SSR-containing genome clone Subtraction library TAC ends TAIL-PCR TDNA tagged Targeting vectors Telomere Associated Sequences U3NeoSV1-trapped U3NeoSV2-trapped XbaI fragments YAC ends cosmid ends cosmid sequence deletion endpoint enhancer trap exon-trapped fosmid ends internal BAC sequence methylation filtered microarray microsatellite paralogous sequence variant partial digestion plasmid plasmid ends plasmid insert plasmid insertion site primer walking random plasmid subclone repeat-enriched representational difference analysis sheared ends shotgun subtractive hybridization transposon insertion site transposon-tagged viral insertion site viral tagged virtual transcript
- The CLASS field is a controlled vocabulary field. Currently (May 2009) accepted values are:
If you have a new category of genome survey sequence to enter, simply email us this information, and we will add it to the accepted list of classes.
- It is important that the strings in the following fields be completely identical:
CONT_NAME of GSS file and NAME field of the Contact file
LIBRARY field of GSS file and NAME field of the Library file
CITATION field of GSS file and TITLE field of the Publication file
These fields from the GSS file are scanned and matched automatically to fields in the Library, Contact, and Publication tables. Differences in content, spelling, letter case, and spacing will result in no match or in an incorrect match.