New FAQ for the BCM Sequence Annotation Server

Randall Smith rsmith at DOT.IMGEN.BCM.TMC.EDU
Tue Sep 13 13:35:54 EST 1994


In response to the feedback we've been getting on our Annotation
Server, and also in response to issues raised by Tom Schneider, Brian
Foley, and Dennis Benson in their recent posts to
bionet.molbio.genbank, we've added a new Overview/FAQ to our Sequence
Annotation server.  The Overview/FAQ describes in more detail the
scope and intended use of our server.  The URL of the new FAQ is:

<a href=http://dot.imgen.bcm.tmc.edu:9331/seq-annot/faq.html> here </a>

A text version of the Overview/FAQ is also appended below (links to
other documents will be missing, so it would be best to view
the URL directly with a WWW client).

All comments, critiques, and suggestions are welcome.

Randall F. Smith, Human Genome Center, Baylor College of Medicine 
rsmith at bcm.tmc.edu

------------------------------------------------------------------------------
           BCM Sequence Annotation Server - Overview and FAQ

Contents:

 1. Notes concerning the purpose and use of this server 
      A. Motivation and Purpose 
      B. Annotations are embedded into sequence records for display purposes
           only 
      C. There are no current plans for annotations to be forwarded to the
           databases 
      D. What happens when an Entrez sequence is updated or corrected by the
           database? 
      E. Originating authors should report updates and corrections directly to
           the appropriate database 
 2. General Information 
      A. Why we chose Entrez as our base sequence database 
      B. Only the annotations are stored on our server 
      C. Annotations cannot be edited or deleted by any user 
 3. Overview of the steps involved in adding a new annotation to the server 
------------------------------------------------------------------------------

1. Notes concerning the purpose and use of this server:

   A. Motivation and Purpose

      Currently there is no easy way for investigators to communicate and
      share new information about nucleic acid and protein sequences. This
      server is an attempt to correct this situation by providing a forum by
      which the research community can easily attach notes to sequence
      records in the Entrez database. 

      We especially hope that users will take advantage of this service to
      share new information about gene/protein function. In order to
      identify a new gene's function using sequence similarity information
      generated during a database search, it is critical that the latest
      functional information be available to the community. However, in a
      large number of cases, functional information is not known at the time
      the sequence is entered into the database. Often it is only from
      subsequent studies (often by third parties) that a gene/protein's
      function is determined. In such cases, it is extremely rare for this
      information to be entered back into a sequence's entry. 

      We hope this server will provide a convenient method for attaching
      new functional information to previously entered sequence records. 

   B. Annotations are embedded into sequence records for display
      purposes only

      Annotations submitted to this server should be viewed as free-text
      electronic "Post-it notes" attached to Entrez sequence reports. The
      annotations are tagged with a "Comment" feature key and embedded
      within sequence records (similar to actual sequence features) for
      display. The annotations should not be considered true features.
      Submitting authors. however, are encouraged to format the
      annotations to appear similar to standard sequence features to enhance
      readability. 

   C. There are no current plans for annotations to be forwarded to the
      databases.

      This server is an attempt to foster the communication of new
      information about sequences. It is not intended to be a mechanism for
      officially updating or correcting the databases. 

   D. What happens when an Entrez sequence is updated or corrected by
      the database?

      All sequences in the Entrez database are assigned a unique identifier,
      the NCBI Seq UID (also known as the "gi" number). When Entrez
      sequences are updated or corrected by the database, if even a single
      base change is made to the sequence, then the the original version of
      the entry is archived by the NCBI and a new Seq UID is assigned to
      the altered sequence. If the update does not involve an change in the
      sequence (e.g., if a feature is added or corrected) then the entry is
      updated without a change in its Seq UID. 

      The NCBI Seq UIDs are used by our server to associate each of
      annotations to the sequence being annotated. When a request to view
      an annotated sequence is made, our server retrieves the most
      up-to-date Entrez sequence report for the Seq UID directly from the 
      NCBI's Entrez WWW Server. The annotations, which are stored
      locally on our server, are then retrieved and embedded into the report
      and displayed. Thus 1) if a database update does not cause a change in
      its Seq UID, the annotations will be automatically attached to the most
      recent entry and 2) if a change in a Seq UID is made then the
      annotations will remain associated with the original sequence entry. 

      In future work we would like to set up an system that will 
      automatically add a notice of change of Seq UID as an annotation to any
      sequence that has been superseded by a new UID. 

   D. Originating authors should report updates and corrections
      directly to the appropriate database.

      The databases will always accept updates and corrections from
      sequence authors. Originating authors should therefore report updates
      directly to the databases. 

      To report updates and corrections to GenBank, either: 
       1. Email a GenBank Update Form (obtained from the 
          NCBI/GenBank Web Page) to: update at ncbi.nlm.nih.gov 

       or

       2. Fill-out the GSDB Online Update Form (this requires a
          WWW client that supports forms). A GenBank update form
          can also be emailed to the GSDB at: update at gsdb.ncgr.org 


2. General Information:

   A. Why we chose Entrez as our base sequence database

      The NCBI Entrez database has a number of features which make it
      extremely useful as our base sequence database: 
      1. The Entrez database is very comprehensive, including nucleic
         acid and protein sequences from most, if not all, of the current
         sequence databases (including GenBank/EMBL/DDBJ, NCBI
         Backbone, PIR, Swiss-Prot, PDB, and PRF) 
      2. The Entrez database employs a stable set of unique identifiers,
         the NCBI Seq UIDs (gi numbers), to reference each sequence in
         the database. 
      3. Sequence reports are available directly over the Internet using
         the NCBI's Entrez WWW Server. 
      4. Since all annotated sequences are linked directly to the the
         NCBI's Entrez WWW Server, the full power of the Entrez
         database's search and retrieval system is available for use
         within our server. This includes, for example, direct access to
         the Medline abstracts of all references included in sequence
         records. 

   B. Only the annotations are stored on our server

      When a user asks to view an annotated sequence, an Entrez sequence
      report is obtained from the NCBI network server. Annotations are
      then retrieved from our local database and embedded in the sequence
      report. 

   C. Annotations cannot be edited or deleted by any user

      If an author of an annotation wishes to edit or delete an annotation,
      please send email to: rsmith at bcm.tmc.edu 



3. Overview of the steps involved in adding a new annotation to the
   server

   Note: To be able to enter an annotation, one must use a WWW client that
   supports forms (e.g., Mosaic 2.x, Chimera, lynx). 

   The steps needed to add a new annotation are outlined here. You should not
   need to refer to this page, however, when adding an annotation, since the 
   steps are described in detail on the individual input forms. 

    1. From the Home Page, select the "Add a new feature annotation to the
       database" line from the list of possible actions. 
    2. On the "Add an Annotation page, enter a user-id if you already have
       one, else leave the field blank and select "Submit" if this is the first
       time you have entered an annotation. 
    3. If you wish to be assigned a new user-id, enter a user-id, then fill-in
       your name, address, etc. on the Personal Information page (this step is
       skipped if you have been previously assigned a user-id). 
    4. Enter either the NCBI Seq UID (gi number) or the accession number
       of the sequence you wish to annotate. If you do not either of these 
       id's, then you can search the Entrez database for a UID or accession 
       number using the NCBI Entrez server. Once you have finished searching
       Entrez, select "Back" from your browser's control panel until you
       return to the Annotation Server. Then enter a Seq UID (preferred) or
       an accession number in the appropriate fields. 
    5. Enter a short annotation into the text field (When the sequence report
       is subsequently viewed from our browser, the annotation will appear
       as a "Comment" feature type embedded in the sequence report). To
       help in preparing the annotation, the Entrez report for the sequence
       being annotated is appended to the bottom of annotation entry page. 
    6. Enter an extended commentary that will provide additional
       information (e.g., literature references) for the annotation. (When
       viewing the sequence report from our browser, the extended
       commentary will be displayed by clicking on the annotation). 
    7. Enter one or more keywords that others can use to find your
       annotation by doing a keyword search of the annotation database. 
    8. Enter one or more URLs that will be attached to the extended
       commentary. This will allow other users to jump to other WWW
       pages that you would liked linked to the annotation (e.g., your own
       home page or publications that you have online). 

------------------------------------------------------------------------------
Randall F. Smith, Human Genome Center, Baylor College of Medicine 
rsmith at bcm.tmc.edu 











More information about the Bio-www mailing list