Note:  This article provides historical information.  For current
       counts of bases and sequences, see the GenBank release notes. 
       
 

	     THE GENBANK DNA SEQUENCE DATABASE

INTRODUCTION

   GenBank* is the NIH's database of all known nucleotide and protein
sequences including supporting bibliographic and biological
information.  As of Release 83.0 in December, 1993, GenBank contained over
163,800,000 nucleotide bases from 150,000 different sequences.  Entries
include a concise description of the sequence, scientific name and
taxonomy of the source organism, and a table of features specifying
coding regions and other sites of biological significance.  As part of
the feature table, protein translations for coding regions are
included.

   GenBank has been the responsibility of the National Center for 
Biotechnology Information (NCBI) since October, 1992.  The NCBI (1) is part of 
the National Library of Medicine (NLM), and, in turn, a part of the National 
Institutes of Health.  Prior to October, 1992, GenBank was funded by the 
National Institute of General Medical Sciences as contracts to IntelliGenetics, Inc. 
(1987-1992), and Bolt, Beranek and Newman, Inc. (1982-1987).  Los Alamos 
National Laboratory (LANL) has participated in GenBank since 1982 as a 
contractor with responsibilty for data entry and maintenance.  An international 
collaboration with the EMBL Data Library in Heidelberg, Germany and the DNA 
Data Bank of Japan (DDBJ) in Mishima provides shared collection and exchange 
of sequence information.

   The history of sequence databases, including GenBank, has been covered 
previously (2,3).  Below are described current operations, recent developments 
and new services provided by NCBI.
			

BUILDING THE DATABASE

   The data in GenBank come from two primary sources:  1) annotators who 
extract the information from relevant journals and, 2) authors who submit data 
directly to the database.  Approximately 36% of the records in GenBank are 
produced by the international collaborators, the EMBL Data Library (32%) and 
DDBJ (4%) with whom sequence information is exchanged on a daily basis.  
NCBI has developed a journal scanning operation in collaboration with Library 
Operations, the NLM division that creates MEDLINE.  A team of annotators 
using NCBI software creates sequence records from journal articles.  Sequence-
containing articles are processed from all of MEDLINE's 3500 journals and from 
an additional set of journals that contain a significant number of sequences.  
Specially trained sequence indexers now have over two years of experience in 
reviewing the literature and creating new sequence entries.  Approximately 9% 
of the current database consists of NCBI-created entries and 15% of the new 
entries are the result of journal scanning.

   With NCBI scanning the literature and sending appropriate information to 
the database collaborators, the international "journal split" is no longer necessary 
and, in fact, the collaborating databases now rely on NCBI to capture published 
sequences.  Another major procedural change is that authors may now send their 
submissions to whatever database is most convenient instead of the database that 
was responsible for the journal in which their article was to be published.  

Direct Submission

   The majority of entries continue to enter the database through direct 
author submission, a process pioneered by LANL over five years ago in response 
to editorial policies that limited the amount of sequence data that were 
publishable in an article reporting those data. .The biological community has 
responded very positively to direct submission and this method has had the 
beneficial effect of involving authors closely in the data input phase of database 
production, thereby increasing the degree of data quality and integrity.  
Submissions are passed back to authors for their review before entering the 
database and are often returned with corrections or updates.  Most importantly, 
authors have become active collaborators and have taken on responsibility for 
the accuracy and completeness of their data.

   Therefore, NCBI strongly encourages authors to submit directly to the 
database prior to publication in order that the sequence can be available to the 
public no later than the time the paper appears in print.  Authors, of course, have 
the right to request that their sequence be kept confidential until the time of 
publication.  In these cases, authors are reminded to inform the database of the 
publication date in order to have a timely release of the data. 

   To help scientists submit sequences and to annotate their data, a program 
called Authorin was developed by IntelliGenetics, Inc. under the previous 
GenBank contract.  It is still available free of charge through the NCBI by 
requesting a copy, preferably by e-mail (authorin@ncbi.nlm.nih.gov) or by 
telephone.  Users should specify whether they prefer the PC or Macintosh 
versions.  NCBI is also developing a platform-independent submission program 
which will enable users to submit sequences over the network and which will be 
available in source code form to enable developers to incorporate the code into 
their software.

ORGANIZATION OF THE DATABASE

   GenBank contains over 129 million bases as of April, 1993, an increase of 
over 46 million bases over the previous 12-month period.  Historically, the 
database has doubled in size about every 22 months.  The traditional flat file is 
distributed in 14 different divisions, such as bacterial, viral, mammalian, primate, 
etc.  Two new divisions were added this year:  PAT for for patent sequences and 
EST for "Expressed Sequence Tags" (see below).  The patent sequences are from 
the U.S. Patent and Trademark Office and from the European Patent Office and 
are being entered into the database as part of an ongoing collaboration between 
the U.S. and European patent offices and the GenBank and EMBL databases, 
respectively.

EST Data

   EST sequences are "single pass," partial DNA sequences that are derived 
from clones that are randomly selected from cDNA libraries and EST data are the 
most rapidly-expanding source of new genes (6).  Because these data differ from 
traditional GenBank entires and thus require special processing and annotation, 
NCBI  also makes them available in a separate database, dbEST, in addition to 
the EST Division of GenBank.  dbEST now includes nearly 20,000 sequences from 
humans, model organisms for genome research, and other species.  NCBI accepts 
bulk submissions of cDNA data and assigns GenBank accession numbers.  Data 
are also accepted from direct submissions to Los Alamos, EMBL and DDBJ.  ESTs 
are automatically screened upon entry and then periodically searched against the 
nucleotide and protein sequence databases in order to identify matches with 
known genes.  The data are stored in a relational database and reformatted as a 
separate (EST) division of the GenBank database.  dbEST sequences can be 
searched by the BLAST e-mail server and full reports of EST records can be 
obtained by querying an e-mail server (est_report@ncbi.nlm.nih.gov).  The full 
reports contain information on the availability of physical cDNA clones and 
mapping data in collaboration with the Genome Data Base at Johns Hopkins .

The Integrated Database (ID)

   In order to produce the GenBank database, NCBI maintains internally an 
Integrated Database, ID, to track and index ASN.1 records from the multiple 
sources of sequence data.  These sources include submissions from LANL, EMBL 
DDBJ, dbEST, and patents plus amino acid sequences from PIR, SWISS-PROT, 
PRF and PDB.  

   ID represents the latest view that each data source has of its data, and 
allows NCBI to assign stable identifiers to sequences.  Through this approach, 
sequence information from a wide variety of sources will have a uniform 
identification system.  These identifiers will be stable and therefore it will be 
easier to know when sequences have changed.  This approach will make the  ID 
database useful as an archive and will also allow scientists to retrace the history 
of revisions for every entry.

   ID will also allow NCBI to reduce unnecessary redundancy in the 
database and thus produce a more useful view of the data for biologists.  
Consider a DNA sequence entry with a coding region and conceptual translation, 
and the corresponding SWISS-PROT entry.  If the conceptual translation is 
identical to the SWISS-PROT entry, it will be replaced with the SWISS-PROT 
entry.  A biologist retrieving the DNA entry will automatically get the full 
annotation of the SWISS-PROT protein.  Likewise, retrieving the SWISS-PROT 
protein would immediately allow one to map to the corresponding DNA 
sequence and coding region.  ID will allow more comprehensive approaches for 
reducing redundancy, and will provide the most biologically relevant entries for 
release.  By maintaining all previous records, it will always be possible for 
scientists to re-evaluate the evolution of the current view of every sequence 
record.

CD-ROM DISTRIBUTION

   The GenBank data is available on CD-ROM through a subscription service 
with the Government Printing Office (202-783-3238, 202-512-2233 FAX).  Order 
forms are also included in each issue of NCBI News, a free subscription to which
may be obtained by contacting NCBI.  NCBI has increased the frequency of major 
releases from four to six per year.  Each release contains a new, full copy of the 
database and is available in the following three versions.

NCBI-GenBank (Flat File)

   This version provides a continuation of the same flat file format in which 
GenBank has been distributed for many years.  Each release is a full release 
incorporating all previous GenBank data supplemented by new data from direct 
submissions, NCBI journal scanning, patents and the EMBL and DDBJ databases. 
Conceptual translations of coding regions appear in feature tables. The release 
contains the standard index files and is organized into divisions.  No
retrieval software is provided. 

Entrez

   Entrez is a two CD-ROM set containing a Sequence disk and a References 
disk.  Entrez: Sequences contains molecular sequence and a set of bibliographic 
references that are cited in the sequence databases.  The DNA and protein 
sequence data are integrated from a variety of sources, including GenBank, 
EMBL, DDBJ, PIR, SWISS-PROT, Protein Research Foundation (PRF),  the 
Brookhaven Protein Data Bank (PDB) and patents.  The second disk contains a 
larger bibliographic subset of MEDLINE(, references to papers indexed under 
the NLM's Medical Subject Heading (MeSH), 'molecular sequence data'.  The 
DNA sequence, protein sequence and bibliographic data are linked to provide 
easy traversal among the databases using a graphical user interface. The retrieval 
system allows for traditional keyword searching and uses pre-computed 
statistical measures of relatedness to allow queries that will find all articles or 
sequences similar to an article or sequence of interest.

   Entrez contains retrieval software for the Apple Macintosh* and for PC-
compatible systems running Microsoft Windows* (version 3.1 or later).  A 
minimum of 2 Mbytes of memory is necessary. Documentation consists of a 30-
page user's guide for installation and operating instructions. (Source code for an 
X11 version of the software for VMS  and Unix platforms is available through 
anonymous FTP from 'ncbi.nlm.nih.gov' in the 'entrez' directory.  Also, 
executables for these and other platforms are available on an unsupported basis.) 

NCBI-Sequences (ASN.1)

   This title provides the integrated sequence dataset used on the Entrez: 
Sequences CD-ROM in the ISO ASN.1 standard data description format.  DNA 
and protein sequence data are linked to journal citations appearing in MEDLINE.  
Files are provided that contain the inter-document/sequence linkage information 
and indices to the byte offsets of the beginning of sequence and bibliographic 
records.  No retrieval software is provided. 

NETWORK ACCESS

Anonymous FTP

   Users on the Internet can use the file transfer protocol (FTP) program to 
download the entire GenBank release or the daily updates.

   Files of the full release and daily updates of the NCBI-GenBank database 
are available for anonymous FTP from 'ncbi.nlm.nih.gov' (130.14.25.1).  The full 
release in flat-file format is available as compressed files in the directory, 'ncbi-
genbank'.  A cumulative update file is contained in the sub-directory, 'daily', and 
a non-cumulative set of updates is in the sub-directory, 'daily-nc'.  ASN.1 
formatted data is in the directory, 'ncbi-asn1'.  Software tools for handling the 
ASN.1 data and for developing ASN.1 applications can be found in the directory 
'toolbox/ncbi_tools'.

E-Mail Servers

   Users with access to electronic mail can search GenBank and eight other 
databases using IRX-based text retrieval.  To start, send a mail message 
containing the word 'help' to: 'retrieve@ncbi.nlm.nih.gov'.  BLAST sequence 
similarity searching (4,5) is also available via e-mail through the address:  
'blast@ncbi.nlm.nih.gov'.  The two e-mail servers are averaging 2000 requests per 
day.

Network Services

   NCBI has begun to offer server-client services on the Internet beginning 
with BLAST clients for the PC, Macintosh, and Unix computers that make direct 
connections to a server at the NCBI.  Over 1200 BLAST requests are processed 
daily through the server-client system.  Client software for Entrez  is undergoing 
testing and will be generally available in the fall, 1993.  Information on 
registering hosts for these clients and obtaining software can be obtained by e-
mail to the address:  'net-info@ncbi.nlm.nih.gov'.   Preliminary reports on the use 
of the client-server approach have been extremely encouraging and support the 
position that this technology offers significant performance and operational 
advantages.  NCBI will make a major commitment to insure that its servers 
maintain the most comprehensive, timely, and accurate version of sequence and 
sequence-related data.  It will also supply stable programming interfaces to 
academic and commercial developers to facilitate the creation of additional 
clients for different platforms.  Other directions planned for server-client services 
include: 1) access to a larger, molecular biology-related subset of MEDLINE, 2) 
greater flexibility in sequence record retrieval, and 3) more powerful similarity 
searching.

MAILING ADDRESS

GenBank
National Center for Biotechnology Information
Bldg. 38A, Rm. 8S-803
8600 Rockville Pike
Bethesda, MD  20894 U.S.A.
301-496-2475

E-MAIL ADDRESSES

info@ncbi.nlm.nih.gov		(general information about NCBI and services)
gb-sub@ncbi.nlm.nih.gov		(submission of sequence data to GenBank)
update@ncbi.nlm.nih.gov		(revisions to GenBank entries and notification 
					of release of 'hold until published' entries)

ACKNOWLEDGEMENTS

	We would like to acknowledge all of the GenBank staff presently at the 
NLM and Los Alamos National Laboratory and all those who have contributed 
to the database since its beginnings over 12 years ago.  The support of Dr.
Ruth Kirchstein and her staff at NIGMS in the first ten years was critical
for its existence.  Also, the GenBank advisors, both national and
international, have contributed their interest and energy towards making the
database a valued resource.

REFERENCES

1. Benson, D., Boguski, M., Lipman, D.J. and Ostell, J. (1990) Genomics, 6, 389-391.

2. Smith, T.F. (1990) Genomics, 6, 702-707.

3. Burks, C., Cinkosky, M.J., Fischer, W.M., Gilna, P., Hayden, J.E.D., Keen, G.M., 
Kelly, M., Kristofferson, D. and Lawrence, J. (1992) Nucleic Acids Research, 20 
(Suppl.), 2065-2069.

4. Altschul, S.F., Gish, W., Miller, W., Myers, E.W. and Lipman, D.J. (1990) J. Mol. 
Biol., 215, 403-410.

5. Gish, W. and States, D.J. (1993) Nature Genetics, 3, 266-272.

6.  Boguski, M.S., Lowe, T.M.J. and Tolstoshev, C.M. (1993) Nature Genetics, 
(1993) 4, 332-3.

*  This description is adapted from: 
 
   Benson, D. Lipman D.J., and Ostell, J. (1993) Nucleic Acids Research, 
   21, 2963-2965.