UniProtKB/TrEMBL - Old Release Notes

N.B. From release 27, the UniProtKB/TrEMBL release notes are included in the UniProt release notes.

TrEMBL Release Notes
Release 10, May 1999

    EMBL Outstation
    European Bioinformatics Institute (EBI)
    Wellcome Trust Genome Campus
    Cambridge CB10 1SD
    United Kingdom

    Telephone: (+44 1223) 494 444
    Fax: (+44 1223) 494 468
    Electronic mail address: DATALIB@EBI.AC.UK
    WWW server:

    Amos Bairoch
    Swiss Institute of Bioinformatics (SIB)
    Centre Medical Universitaire
    1, rue Michel Servet
    1211 Geneva 4

    Telephone: (+41 22) 784 40 82
    Fax: (+41 22) 702 55 02
    Electronic mail address: BAIROCH@CMU.UNIGE.CH
    WWW server:


    TrEMBL has been prepared by:

    o  Rolf Apweiler, Kirsty Bates, Margaret Biswas, Sergio Contrino, 
       Wolfgang Fleischmann, Gill Fraser, Henning Hermjakob, Vivien Junker,
       Youla Karavidopoulou, Fiona Lang,  Minna Lehvaslaiho, Michele Magrane,
       Maria Jesus Martin, Steffen Moeller, Nicoletta Mitaritonna, Nicola Mulder,
       Claire O'Donovan and Eleanor Whitfield
       at the EMBL Outstation - European Bioinformatics Institute (EBI) in
       Hinxton, UK;
    o  Amos Bairoch and Alain Gateau at the Swiss Institute of Bioinformatics
       in Geneva, Switzerland.


    This manual and the database it accompanies may be copied and
    redistributed freely, without advance permission, provided
    that this statement is reproduced with each copy.


    If you  want to  cite  TrEMBL  in  a  publication  please  use  the
    following reference:

              Bairoch A, and Apweiler R.
              The Swiss-Prot protein sequence data bank and its supplement
              TrEMBL in 1999.
              Nucleic Acids Res. 27:49-54(1999).

                         1. Introduction

TrEMBL is a computer-annotated protein sequence database supplementing the
Swiss-Prot Protein Sequence Data Bank. TrEMBL contains the translations of
all coding sequences (CDS) present in the EMBL Nucleotide Sequence Database
not yet integrated in Swiss-Prot. TrEMBL can be considered as a preliminary
section of Swiss-Prot. For all TrEMBL entries which should finally be
upgraded to the standard Swiss-Prot quality, Swiss-Prot accession numbers
have been assigned.

                        2. Why a supplement to Swiss-Prot?

The ongoing gene sequencing and mapping projects have dramatically
increased the number of protein sequences to be incorporated into Swiss-Prot.
We do not want to dilute the quality standards of Swiss-Prot by incorporating
sequences without proper sequence analysis and annotation, but we do want to
make the sequences available as fast as possible. TrEMBL achieves this second
goal, and is a major step in the process of speeding up subsequent
upgrading of annotation to the standard Swiss-Prot quality.
To address the problem of redundancy, the translations of all coding
sequences (CDS) in the EMBL Nucleotide Sequence Database already included
in Swiss-Prot have been removed from TrEMBL.

We name this supplement TrEMBL (Translation from EMBL), since the tools
used to create the translations of the CDS are based on the program
'trembl' written by Thure Etzold at the EMBL.

                             3. The Release

This TrEMBL release was created from the EMBL Nucleotide Sequence Database
release 58 and contains 244'862 sequence entries, comprising 66'562'800 amino
acids. To minimise redundancy, the translations of all coding sequences (CDS)
in the EMBL Nucleotide Sequence Database already included in Swiss-Prot 37 
have been removed from TrEMBL release 10.

TrEMBL is split in two main sections: SP-TrEMBL and REM-TrEMBL:

SP-TrEMBL (Swiss-Prot TrEMBL) contains the entries (201'082) which should be
eventually incorporated into Swiss-Prot. Swiss-Prot accession numbers have 
been assigned for all SP-TrEMBL entries.

SP-TrEMBL is organised in subsections:

arc.dat (Archea):               7408 entries
fun.dat (Fungi):                6679 entries
hum.dat (Human):                8518 entries
inv.dat (Invertebrates):       23653 entries
mam.dat (Other Mammals):        3130 entries
mhc.dat (MHC proteins):         4236 entries
org.dat (Organelles):          16261 entries
phg.dat (Bacteriophages):       1971 entries
pln.dat (Plants):              17352 entries
pro.dat (Prokaryotes):         45992 entries
rod.dat (Rodents):              7480 entries
unc.dat (Unclassified):           44 entries
vrl.dat (Viruses):             53916 entries
vrt.dat (Other Vertebrates):    4442 entries

25166 new entries have been integrated in SP-TrEMBL. The sequences of 
894 SP-TrEMBL entries have been updated and the annotation has been updated in 
57'510 entries.

In the document deleteac.txt, you will find a list of all accession numbers 
which were previously present in TrEMBL, but which have now been deleted from 
the database.

REM-TrEMBL (REMaining TrEMBL) contains the entries (43'780) that we do
not want to include in Swiss-Prot. REM-TrEMBL entries have no accession
numbers. This section is organised in five subsections:

   1) Immunoglobulins and T-cell receptors (Immuno.dat)
      Most REM-TrEMBL entries are  immunoglobulins and  T-cell receptors. We
      stopped entering immunoglobulins and T-cell receptors into Swiss-Prot,
      because we only want to keep  the  germ line gene derived translations 
      of these proteins in Swiss-Prot and not all known somatic recombinated
      variations of  these proteins.  We would like to  create a specialised 
      database  dealing  with  these  sequences  as a further  supplement to 
      Swiss-Prot  and  keep  only a  representative  cross-section of  these 
      proteins in Swiss-Prot.

   2) Synthetic sequences (Synth.dat)
      Another category of data, which will not be included in Swiss-Prot are
      synthetic sequences.  Again, we do not want to  leave these entries in
      TrEMBL. Ideally one should build a specialised database for artificial 
      sequences as a further supplement to Swiss-Prot.

   3) Patent application sequences (Patent.dat)
      A third  subsection consists of  coding sequences captured from patent
      applications.  A thorough  survey of  these  entries  have shown  that 
      apart from a rather small minority  (which in  most cases have already 
      been integrated in Swiss-Prot), most of these sequences contain either
      erroneous data or concern artificially generated sequences outside the 
      scope of Swiss-Prot.

   4) Small fragments (Smalls.dat)
      Another  subsection  consists of fragments  with less than eight amino

   5) CDS not coding for real proteins (Pseudo.dat)
      The last subsection consists of  CDS translations where we have strong
      evidence to believe that these CDS are not coding for real proteins.

                4. Format Differences Between Swiss-Prot and TrEMBL

The format and conventions used by TrEMBL follow as closely as possible
that of Swiss-Prot. Hence, it is not necessary to produce an additional
user manual and extensive release notes for TrEMBL. The information given
in the Swiss-Prot release notes and user manual are in general valid for
TrEMBL. The differences are mentioned below.

The general structure of an entry is identical in Swiss-Prot and TrEMBL.
The data class used in TrEMBL (in the ID line) is always 'PRELIMINARY',
whereas in Swiss-Prot it is always 'STANDARD'.

Differences in line types present in Swiss-Prot and TrEMBL:

The ID line (IDentification):

The entry name used in SP-TrEMBL is the same as the Accession Number of the
entry. The entry name used in REM-TrEMBL is the protein_id tagged to the
corresponding CDS in the EMBL Nucleotide Sequence Database. 'protein_id' 
stands for the "Protein IDentification" number. It is a number that you 
will find in the feature table of the EMBL nucleotide sequence entries 
in a qualifier called "/protein_id" which is tagged to every CDS.


FT   CDS             339..1514
FT                   /codon_start=1
FT                   /db_xref="PID:g1256015"
FT                   /product="dystrobrevin-epsilon"
FT                   /protein_id="AAC50431.1"

The DT line (DaTe)

The format of the DT lines that serve to indicate when an entry was
created and updated are identical to that defined in Swiss-Prot; but the
DT lines in TrEMBL are referring to the TrEMBL release. The difference is
shown in the example below.

    DT lines in a Swiss-Prot entry:

    DT   01-JAN-1988 (Rel. 06, Created)
    DT   01-JUL-1989 (Rel. 11, Last sequence update)
    DT   01-AUG-1992 (Rel. 23, Last annotation update)

    DT lines in a TrEMBL entry:

    DT   01-NOV-1996 (TrEMBLrel. 01, Created)
    DT   01-NOV-1996 (TrEMBLrel. 01, Last sequence update)
    DT   01-FEB-1997 (TrEMBLrel. 02, Last annotation update)

                5. Weekly updates of TrEMBL and non-redundant data sets

Weekly cumulative updates of TrEMBL are available by anonymous FTP and
from the EBI SRS server.

We also produce every week a complete non-redundant protein sequence 
collection by providing three compressed files (these are in the directory 
/pub/databases/sp_tr_nrdb on the EBI FTP server and in databases/sp_tr_nrdb
on the ExPASy server): sprot.dat.Z, trembl.dat.Z and trembl_new.dat.Z.
This set of non-redundant files is especially important for two types of 
(i) Managers of similarity search services. They can now provide what is 
currently the most comprehensive and non-redundant data set of protein 
(ii) Anybody wanting to update their full copy of Swiss-Prot + TrEMBL to 
their own schedule without having to wait for full releases of Swiss-Prot
or of TrEMBL.

                6. Access/Data Distribution

FTP server:
SRS server:

TrEMBL is also available on the Swiss-Prot CD-ROM.
Swiss-Prot + TrEMBL is searchable on the following servers at the EBI:
Bic_sw  (
Scanps  (
SSearch (

                7. Description of changes made to TrEMBL since release 9.

7.1  Extension of the accession number system.

As all possible numbers with the 'O' series have been used, a new system of 
accession numbers has been introduced in TrEMBL release 10. This system is 
based on the following format:

    1        2       3          4            5            6
    [O,P,Q]  [0-9]  [A-Z, 0-9]  [A-Z, 0-9]   [A-Z, 0-9]   [0-9]

What the above means is that we will keep a six-character code, but that
in positions 3, 4 and 5 of this code a letter or a number can be present.  
We allow only numbers in positions 2 and 6. 

Examples: Q9ZXM8, Q9ZNQ2, Q9YGB9, Q9YH55

7.2  New Protein Sequence Identifier in the cross-references to the EMBL 
nucleotide sequence database.

The new protein sequences identifiers introduced in the EMBL nucleotide
sequence database release 58 are now present in all cross-references
to the EMBL nucleotide sequence database and replace the previous PIDs.

The new protein sequence identifier consists of a stable ID portion 
(3+5 format with 3 position letters and 5 numbers) plus a version number 
after a decimal point.  

DR   EMBL; L37685; AAC41668.1; -.

7.3  Conversion of TrEMBL to mixed cases.

In this release we have converted to mixed cases the following line

        DT, OS, OG, OC, RL and KW

This conversion is described in detail in section 3.1 of the release notes
of Swiss-Prot release 37.0.

7.4. RL lines.

The format for RL lines for submissions to DNA databases has changed.

Example: in release 9 the format was:


In release 10 it is :

RL   Submitted (APR-1996) to the EMBL/GenBank/DDBJ databases.