Greetings GenBank Users,
GenBank Release 199.0 is now available via FTP from the
National Center for Biotechnology Information (NCBI):
Ftp Site Directory Contents
---------------- --------- ---------------------------------------
ftp.ncbi.nih.gov genbank GenBank Release 199.0 flatfiles
ncbi-asn1 ASN.1 data used to create Release 199.0
Close-of-data for GenBank 199.0 occurred on 12/10/2013. Uncompressed,
the Release 199.0 flatfiles require roughly 618 GB (sequence files only).
The ASN.1 data require approximately 508 GB.
Recent statistics for non-WGS, non-CON sequences:
Release Date Base Pairs Entries
198 Oct 2013 155176494699 168335396
199 Dec 2013 156230531562 169331407
Recent statistics for WGS sequences:
Release Date Base Pairs Entries
198 Oct 2013 535842167741 130203205
199 Dec 2013 556764321498 133818570
During the 54 days between the close dates for GenBank Releases 198.0
and 199.0, the non-WGS/non-CON portion of GenBank grew by 1,054,036,863
basepairs and by 996,011 sequence records. During that same period,
494,249 records were updated. An average of 27,597 non-WGS/non-CON
records were added and/or updated per day.
Between releases 198.0 and 199.0, the WGS component of GenBank grew by
20,922,153,757 basepairs and by 3,615,365 sequence records.
For additional release information, see the README files in either of
the directories mentioned above, and the release notes (gbrel.txt) in
the genbank directory. Sections 1.3 and 1.4 of the release notes
(Changes in Release 199.0 and Upcoming Changes) have been appended
below for your convenience.
* * * Important * * *
Significant changes are described in Sections 1.3.2 and 1.4.1 of the
release notes: A new accession number format for WGS scaffolds, and
an anticipated phasing-out of NCBI GI sequence identifiers. Users who
make use of GIs in their information systems and analysis pipelines
should take particular note of section 1.4.1.
Release 199.0 data, and subsequent updates, are available now via
NCBI's Entrez and Blast services.
As a general guideline, we suggest first transferring the GenBank
release notes (gbrel.txt) whenever a release is being obtained. Check
to make sure that the date and release number in the header of the
release notes are current (eg: December 15 2013, 199.0). If they are
not, interrupt the remaining transfers and then request assistance from
the NCBI Service Desk.
A comprehensive check of the headers of all release files after your
transfers are complete is also suggested. Here's how one might go about
this on a Unix or Linux platform, using csh/tcsh :
set files = `ls gb*.*`
foreach i ($files)
head -10 $i | grep Release
end
Or, if the files are compressed, perhaps:
gzcat $i | head -10 | grep Release
If you encounter problems while ftp'ing or uncompressing Release
199.0, please send email outlining your difficulties to:
info from ncbi.nlm.nih.gov
Mark Cavanaugh, Michael Kimelman, Ilya Dondoshansky, Sergey Zhdanov,
GenBank
NCBI/NLM/NIH/HHS
1.3 Important Changes in Release 199.0
1.3.1 Organizational changes
The total number of sequence data files increased by 18 with this release:
- the BCT division is now composed of 114 files (+2)
- the CON division is now composed of 231 files (+5)
- the ENV division is now composed of 67 files (+2)
- the EST division is now composed of 475 files (+1)
- the GSS division is now composed of 279 files (+1)
- the PAT division is now composed of 199 files (+2)
- the PLN division is now composed of 65 files (+1)
- the TSA division is now composed of 147 files (+2)
- the VRL division is now composed of 29 files (+2)
1.3.2 New accession format for CON-division WGS scaffold records
WGS scaffolds that are constructed from WGS contigs have
traditionally made use of a '2+6' accession number format,
with two leading alphabetic characters followed by six digits.
Here is an example of a WGS-master record that references two
different ranges of scaffold accession numbers:
http://www.ncbi.nlm.nih.gov/nuccore/AABR00000000
LOCUS AABR06000000 112651 rc DNA linear ROD 16-MAR-2012
DEFINITION Rattus norvegicus strain BN/SsNHsdMCW, whole genome shotgun
sequencing project.
ACCESSION AABR00000000
VERSION AABR00000000.6 GI:380236478
DBLINK BioProject: PRJNA10629
KEYWORDS WGS.
SOURCE Rattus norvegicus (Norway rat)
ORGANISM Rattus norvegicus
....
....
WGS AABR06000001-AABR06112651
WGS_SCAFLD CM000072-CM000092
WGS_SCAFLD JH612139-JH620698
//
Many WGS projects have a large number of chromosome-specific scaffolds
(such as the JH accession range for AABR), and a much smaller number of
scaffolds that represent the entirety of the chromosomes (such as the the CM
accession range for AABR). Because of the former, we are consuming '2+6'
prefixes, like JH, at an unsustainable rate.
So as of the December 2013 GenBank release, a new accession format for
WGS scaffolds which mirrors the format of the underlying WGS contigs will
begin to appear for newly-processed WGS projects:
4 letter WGS project code
2 digit assembly-version number
"S" (for 'scaffold')
Six or seven digits
In the above example, the set of 'JH' scaffolds could make use of
accession numbers such as AABR06S000001 and AABR06S112651 :
WGS AABR06000001-AABR06112651
WGS_SCAFLD CM000072-CM000092
WGS_SCAFLD AABR06S000001-AABR06S112651
We do not currently plan to replace existing '2+6' accessions with the new
'4+2+S+6/7' accessions.
This new accession format will initially be used for WGS projects that :
a) Have a very large number of contigs (typically, greater than 1 million)
b) Have a correspondingly large number of scaffolds
c) Are completely unannotated, at both the contig and scaffold level.
The first WGS project which has these properties is ALWZ02. Here's an
excerpt of the flatfile for the first ALWZ scaffold, which illustrates
the new accession number format:
LOCUS ALWZ02S0000001 701 bp DNA linear CON 14-JUN-2013
DEFINITION Picea glauca scaffold316, whole genome shotgun sequence.
ACCESSION ALWZ02S0000001 ALWZ000000000
VERSION ALWZ02S0000001.1
DBLINK BioProject: PRJNA83435
KEYWORDS WGS.
SOURCE Picea glauca (white spruce)
URLs at which these scaffolds can be obtained include:
http://www.ncbi.nlm.nih.gov/Traces/wgs/?val=ALWZ02#scaffoldshttp://www.ncbi.nlm.nih.gov/Traces/wgs/?download=ALWZ02S.gbff.1.gzftp://ftp.ncbi.nlm.nih.gov/genbank/wgs/wgs.ALWZ.scflds.1.gbff.gz
1.3.3 Support for higher-precision /collection_date values
At the May 2013 meeting of the International Nucleotide Sequence Database
Collaboration, it was agreed to provide support for collection dates that
include the time, and for collection date ranges.
The new format for /collection_date is theoretically legal as of this
GenBank Release 199.0, of December 2013. However, given that a large
number of submission tools and systems will have to be updated, it's
possible that the new date/time format will not actually be encountered
by users until 2014.
Here is the working draft of the new /collection_date definition:
Qualifier /collection_date=
Definition The date on which the specimen was collected.
Date/time ranges are supported by providing two collection
dates from among the supported value formats, delimited by a
forward-slash character.
Collection times are supported by adding "T", then the hour
and minute, after the date. Collection times must be in
Coordinated Universal Time (UTC), otherwise known as
"Zulu Time" (Z).
Value format "DD-Mmm-YYYY", "Mmm-YYYY", "YYYY"
YYYY-MM-DDThh:mmZ, YYYY-MM-DDThhZ, YYYY-MM-DD, or YYYY-MM
Example /collection_date="21-Oct-1952"
/collection_date="Oct-1952"
/collection_date="1952"
/collection_date="1952-10-21T11:43Z"
/collection_date="1952-10-21T11Z"
/collection_date="1952-10-21"
/collection_date="1952-10"
/collection_date="21-Oct-1952/15-Feb-1953"
/collection_date="Oct-1952/Feb-1953"
/collection_date="1952/1953"
/collection_date="1952-10-21/1953-2-15"
/collection_date="1952-10/1953-2"
/collection_date="1952-10-21T11:43Z/1952-10-21T17:43Z"
Comment 'Mmm' represents a three-letter month abbreviation, and can
be one of the following:
Jan, Feb, Mar, Apr, May, Jun, Jul, Aug, Sep, Oct, Nov, Dec
'YYYY' is a four-digit value representing the year. 'MM' is
a two-digit value representing the month. 'DD' is a two-digit
value representing the day of the month.
'hh' is a two-digit value representing the hour of the day
(00 to 23)
'mm' is a two-digit value representing the minute of the hour
(01 to 59)
Within a date range, value formats that make use of 'Mmm'
(month abbreviations) cannot be combined with value formats
that make use of 'MM' (two-digit month number).
Collection dates that are specified to at least the month,
day, and year (DD-Mmm-YYYY or YYYY-MM-DD) are strongly
encouraged. If the day and/or month of the collection date
are not known, Mmm-YYYY or YYYY-MM or YYYY may be used.
Within a collection date range, the first date (possibly
including time) must be prior to the second date (possibly
including time).
Within a collection date range for which the day, month,
and year are identical, the first time value must be prior
to the second time value.
1.3.4 GSS File Header Problem
GSS sequences at GenBank are maintained in two different systems, depending
on their origin, and the dumps from those systems occur in parallel. Because
the second dump (for example) has no prior knowledge of exactly how many GSS
files will be dumped by the first, it does not know how to number its own
output files.
There is thus a discrepancy between the filenames and file headers for 118
of the GSS flatfiles in Release 199.0. Consider gbgss162.seq :
GBGSS1.SEQ Genetic Sequence Data Bank
December 15 2013
NCBI-GenBank Flat File Release 199.0
GSS Sequences (Part 1)
87041 loci, 63875275 bases, from 87041 reported sequences
Here, the filename and part number in the header is "1", though the file
has been renamed as "162" based on the number of files dumped from the other
system. We hope to resolve this discrepancy at some point, but the priority
is certainly much lower than many other tasks.
1.4 Upcoming Changes
1.4.1 GI sequence identifiers to be phased out (slowly!) at NCBI
The numeric GI sequence identifier that NCBI assigns to all nucleotide
and protein sequences was first introduced for GenBank Release products
as of GenBank 81.0, in February 1994. See:
ftp://ftp.ncbi.nih.gov/genbank/release.notes/gb81.release.notes
These simple, uniform, integer-based unique identifiers (which predated the
introduction of Accession.Version sequence identifiers) were crucial to the
development of NCBI's Entrez retrieval system, and have served their purpose
very well for nearly 20 years.
However, as NCBI considers how best to address the expected increase in the
volume of submitted sequence data, it is clear that prior practices will need
to be re-thought. As an example, imagine 100,000 pathogen-related
genomes/samples, each with 5000 proteins, most of which are common to all. We
will be moving toward solutions that represent each unique protein *once*.
The coding region protein products for each genome will likely continue to be
assigned their own Accession.Version identifiers, but (within the NCBI data
model) they will simply *reference* the unique proteins. And, they will no
longer be issued GIs of their own.
Such a change will likely have a significant impact on NCBI users who utilize
GIs in their own information systems and analysis pipelines, so it will not be
introduced quickly. You can expect that a great deal of additional detail will
be made available via NCBI's various announcement mechanisms.
*This* particular announcement is chiefly intended to provide some advance
warning to our users. There _will_ be classes of GenBank sequences that
are not assigned GIs in the not-too-distant future. If GIs are central to
your operations, then it might be appropriate to begin planning a switch to
the use of Accession.Version identifiers instead.
And in fact, NCBI now has at least one WGS submission for which GIs have
not been assigned, for both the contigs and the scaffolds : ALWZ02.
Here are excerpts of the flatfile representation for the first ALWZ02 contig,
and the 'singleton' scaffold which is constructed from it:
LOCUS ALWZ020000001 701 bp DNA linear PLN 28-MAY-2013
DEFINITION Picea glauca contig316_0, whole genome shotgun sequence.
ACCESSION ALWZ020000001 ALWZ020000000
VERSION ALWZ020000001.1
DBLINK BioProject: PRJNA83435
....
ORIGIN
1 cgttgtgttg gggcacccaa ccttggtgag gccgtattaa aaagtctacc tccaagccaa
61 aatttgttct tatccatcct ccaactcgtc tttttgccta gtgctcccct atgtggacgt
121 tttcgttgtg gaggagtttt tcgtttgggc gcccatcttg cgaactcacc ttgcattgcg
181 tttggtcgcc caacttgtga acgtgccttg gattgcgttg gggcgcccaa gttgcggacg
241 tgcggacgtg cctttctttg ccgacatgcc ttgcgtttgc gttgcggacg tgcaatgggg
301 cccccagctt gctgacgtgc cttgcgttgc gttggggtgc ccaacttgcc gacgtgcctt
361 gcgttgcgtt ggggcaccca accttggtga ggccgtatta aaaagtctac ctccaagata
421 aaatttgttc ttatccatcc tctagctctt cttttagcct agtgctccct tgtgtggaca
481 ttttcgttgt ggatcatttt ttcgtttagg tgcccatctt gcagacaagc cttgcgttgc
541 gtttgggcac ccatcttgcg gacgcgcctt tcattgcgtt ggggcgccca acgttggtga
601 ggccgtatta aaaagtctac ctccaagaca aaatttgttc ttatccatcc tccatctcgt
661 ctttttgcct agtgctccct tgtgtggacg ttttcgttgc g
//
LOCUS ALWZ02S0000001 701 bp DNA linear CON 14-JUN-2013
DEFINITION Picea glauca scaffold316, whole genome shotgun sequence.
ACCESSION ALWZ02S0000001 ALWZ000000000
VERSION ALWZ02S0000001.1
DBLINK BioProject: PRJNA83435
....
CONTIG join(ALWZ020000001.1:1..701)
//
**Note the absence of a GI value on the VERSION line of these two records**
Sample URLs from which ALWZ02 data may be obtained include:
http://www.ncbi.nlm.nih.gov/Traces/wgs/?val=ALWZ02#contigshttp://www.ncbi.nlm.nih.gov/Traces/wgs/?val=ALWZ02#scaffoldshttp://www.ncbi.nlm.nih.gov/Traces/wgs/?download=ALWZ02.gbff.1.gzhttp://www.ncbi.nlm.nih.gov/Traces/wgs/?download=ALWZ02S.gbff.1.gzftp://ftp.ncbi.nlm.nih.gov/genbank/wgs/wgs.ALWZ.1.gbff.gzftp://ftp.ncbi.nlm.nih.gov/genbank/wgs/wgs.ALWZ.scflds.1.gbff.gz
Unannotated WGS projects consisting of many millions of contigs and
scaffolds represent the first class of records for which GIs are no
longer being assigned. ALWZ02 is the only such project at this time,
but more projects will follow, and the practice will ultimately expand
to include other classes of records.
