Menu
▾
▴
gusdev-gusdev
|
From: Angel P. <an...@sn...> - 2002-07-26 15:09:32
|
I want to add a table BibliographicReference to GUS30 SRes, in addition to the Reference table, or at least revamp the Reference table by adding the missing columns that are in BibRef (attached below). The reasoning is that the Reference table is a hold-over from parsing genbank records for this type of info. Thus the position column refers to the positional information from genbank features. Also we have the old 'remark' column from Jian Wang's days of schema design. Here is the schema def for BibRef: CREATE TABLE SRes.BibliographicReference ( BIBLIOGRAPHIC_REFERENCE_ID NUMBER(10) NOT NULL, TITLE VARCHAR2 ( 100 ) NOT NULL, AUTHORS VARCHAR2 ( 100 ) NOT NULL, PUBLICATION VARCHAR2 ( 100 ), EDITOR VARCHAR2 ( 100 ), YEAR NUMBER(5), VOLUME VARCHAR2 ( 100 ), ISSUE VARCHAR2 ( 100 ), PAGES VARCHAR2 ( 100 ), URI VARCHAR2 ( 100 ), MODIFICATION_DATE DATE NOT NULL, USER_READ NUMBER ( 1 ) NOT NULL, USER_WRITE NUMBER ( 1 ) NOT NULL, GROUP_READ NUMBER ( 1 ) NOT NULL, GROUP_WRITE NUMBER ( 1 ) NOT NULL, OTHER_READ NUMBER ( 1 ) NOT NULL, OTHER_WRITE NUMBER ( 1 ) NOT NULL, ROW_USER_ID NUMBER ( 12 ) NOT NULL, ROW_GROUP_ID NUMBER ( 12 ) NOT NULL, ROW_PROJECT_ID NUMBER ( 12 ) NOT NULL, ROW_ALG_INVOCATION_ID NUMBER ( 12 ) NOT NULL, CONSTRAINT PK_BIBLIOGRAPHICREFERNECE PRIMARY KEY (BIBLIOGRAPHIC_REFERENCE_ID) ); |
|
From: Arnaud K. <ax...@sa...> - 2002-07-29 16:11:33
|
Hi Angel
some comments:
Could it be possible to add the following fields:
* abstract,
* pubmed_id,
* medline_id,
* is_curated,
* the list of keywords, in a separate table, ReferenceKeyword.
Also, we would like to store each author in a separate entry instead of
a unique column.
Besides, to facilitate the queries, it would be better to have a many to
many relationships between BibliographicReference and the two tables,
ReferenceKeyword and Author.
cheers
Arnaud
Angel Pizarro wrote:
>I want to add a table BibliographicReference to GUS30 SRes, in addition to
>the Reference table, or at least revamp the Reference table by adding the
>missing columns that are in BibRef (attached below).
>
>The reasoning is that the Reference table is a hold-over from parsing
>genbank records for this type of info. Thus the position column refers to
>the positional information from genbank features. Also we have the old
>'remark' column from Jian Wang's days of schema design. Here is the schema
>def for BibRef:
>
>CREATE TABLE SRes.BibliographicReference (
> BIBLIOGRAPHIC_REFERENCE_ID NUMBER(10) NOT NULL,
> TITLE VARCHAR2 ( 100 ) NOT NULL,
> AUTHORS VARCHAR2 ( 100 ) NOT NULL,
> PUBLICATION VARCHAR2 ( 100 ),
> EDITOR VARCHAR2 ( 100 ),
> YEAR NUMBER(5),
> VOLUME VARCHAR2 ( 100 ),
> ISSUE VARCHAR2 ( 100 ),
> PAGES VARCHAR2 ( 100 ),
> URI VARCHAR2 ( 100 ),
> MODIFICATION_DATE DATE NOT NULL,
> USER_READ NUMBER ( 1 ) NOT NULL,
> USER_WRITE NUMBER ( 1 ) NOT NULL,
> GROUP_READ NUMBER ( 1 ) NOT NULL,
> GROUP_WRITE NUMBER ( 1 ) NOT NULL,
> OTHER_READ NUMBER ( 1 ) NOT NULL,
> OTHER_WRITE NUMBER ( 1 ) NOT NULL,
> ROW_USER_ID NUMBER ( 12 ) NOT NULL,
> ROW_GROUP_ID NUMBER ( 12 ) NOT NULL,
> ROW_PROJECT_ID NUMBER ( 12 ) NOT NULL,
> ROW_ALG_INVOCATION_ID NUMBER ( 12 ) NOT NULL,
> CONSTRAINT PK_BIBLIOGRAPHICREFERNECE PRIMARY KEY
>(BIBLIOGRAPHIC_REFERENCE_ID)
>);
>
>
>
>
|
|
From: Angel P. <an...@sn...> - 2002-07-29 20:05:09
|
Arnaud, this is a bit over powered for what we use, but I think it not all that much more, except the bit about the authors. Can I make the compromise to keep the author field in the table in addition to a normalized set of tables for those that choose to use them? ALso I had forgotten to add an (external_db_release_id, source_id) tuple to the table before I sent it out. This would only take care of one external reference, though. A possible solution would be to use a many to many relation to DbRef for all of the external ids such as pubmed and medline. Lastly, what is your intent with the 'is_curated' bit column? Angel On Mon, 29 Jul 2002, Arnaud Kerhornou wrote: > Hi Angel > > some comments: > > Could it be possible to add the following fields: > * abstract, > * pubmed_id, > * medline_id, > * is_curated, > * the list of keywords, in a separate table, ReferenceKeyword. > > Also, we would like to store each author in a separate entry instead of > a unique column. > > Besides, to facilitate the queries, it would be better to have a many to > many relationships between BibliographicReference and the two tables, > ReferenceKeyword and Author. > > cheers > Arnaud > > Angel Pizarro wrote: > > >I want to add a table BibliographicReference to GUS30 SRes, in addition to > >the Reference table, or at least revamp the Reference table by adding the > >missing columns that are in BibRef (attached below). > > > >The reasoning is that the Reference table is a hold-over from parsing > >genbank records for this type of info. Thus the position column refers to > >the positional information from genbank features. Also we have the old > >'remark' column from Jian Wang's days of schema design. Here is the schema > >def for BibRef: > > > >CREATE TABLE SRes.BibliographicReference ( > > BIBLIOGRAPHIC_REFERENCE_ID NUMBER(10) NOT NULL, > > TITLE VARCHAR2 ( 100 ) NOT NULL, > > AUTHORS VARCHAR2 ( 100 ) NOT NULL, > > PUBLICATION VARCHAR2 ( 100 ), > > EDITOR VARCHAR2 ( 100 ), > > YEAR NUMBER(5), > > VOLUME VARCHAR2 ( 100 ), > > ISSUE VARCHAR2 ( 100 ), > > PAGES VARCHAR2 ( 100 ), > > URI VARCHAR2 ( 100 ), > > MODIFICATION_DATE DATE NOT NULL, > > USER_READ NUMBER ( 1 ) NOT NULL, > > USER_WRITE NUMBER ( 1 ) NOT NULL, > > GROUP_READ NUMBER ( 1 ) NOT NULL, > > GROUP_WRITE NUMBER ( 1 ) NOT NULL, > > OTHER_READ NUMBER ( 1 ) NOT NULL, > > OTHER_WRITE NUMBER ( 1 ) NOT NULL, > > ROW_USER_ID NUMBER ( 12 ) NOT NULL, > > ROW_GROUP_ID NUMBER ( 12 ) NOT NULL, > > ROW_PROJECT_ID NUMBER ( 12 ) NOT NULL, > > ROW_ALG_INVOCATION_ID NUMBER ( 12 ) NOT NULL, > > CONSTRAINT PK_BIBLIOGRAPHICREFERNECE PRIMARY KEY > >(BIBLIOGRAPHIC_REFERENCE_ID) > >); > > > > > > > > > > |
|
From: Arnaud K. <ax...@sa...> - 2002-07-30 15:10:51
|
Hi Angel Angel Pizarro wrote: >Arnaud, >this is a bit over powered for what we use, but I think it not all >that much more, except the bit about the authors. Can I make the >compromise to keep the author field in the table in addition to a >normalized set of tables for those that choose to use them? > Do you mean you want to keep the authors column in addition of a table authors ? If so it's fine. > >ALso I had forgotten to add an (external_db_release_id, source_id) tuple >to the table before I sent it out. This would only take care of one >external reference, though. A possible solution would be to use a many to >many relation to DbRef for all of the external ids such as pubmed and >medline. > or the ExternalDatabaseLink table. > >Lastly, what is your intent with the 'is_curated' bit column? >Angel > Val is going to answer to your question. cheers Arnaud > >On Mon, 29 Jul 2002, Arnaud Kerhornou wrote: > >>Hi Angel >> >>some comments: >> >>Could it be possible to add the following fields: >> * abstract, >> * pubmed_id, >> * medline_id, >> * is_curated, >> * the list of keywords, in a separate table, ReferenceKeyword. >> >>Also, we would like to store each author in a separate entry instead of >>a unique column. >> >>Besides, to facilitate the queries, it would be better to have a many to >>many relationships between BibliographicReference and the two tables, >>ReferenceKeyword and Author. >> >>cheers >>Arnaud >> >>Angel Pizarro wrote: >> >>>I want to add a table BibliographicReference to GUS30 SRes, in addition to >>>the Reference table, or at least revamp the Reference table by adding the >>>missing columns that are in BibRef (attached below). >>> >>>The reasoning is that the Reference table is a hold-over from parsing >>>genbank records for this type of info. Thus the position column refers to >>>the positional information from genbank features. Also we have the old >>>'remark' column from Jian Wang's days of schema design. Here is the schema >>>def for BibRef: >>> >>>CREATE TABLE SRes.BibliographicReference ( >>> BIBLIOGRAPHIC_REFERENCE_ID NUMBER(10) NOT NULL, >>> TITLE VARCHAR2 ( 100 ) NOT NULL, >>> AUTHORS VARCHAR2 ( 100 ) NOT NULL, >>> PUBLICATION VARCHAR2 ( 100 ), >>> EDITOR VARCHAR2 ( 100 ), >>> YEAR NUMBER(5), >>> VOLUME VARCHAR2 ( 100 ), >>> ISSUE VARCHAR2 ( 100 ), >>> PAGES VARCHAR2 ( 100 ), >>> URI VARCHAR2 ( 100 ), >>> MODIFICATION_DATE DATE NOT NULL, >>> USER_READ NUMBER ( 1 ) NOT NULL, >>> USER_WRITE NUMBER ( 1 ) NOT NULL, >>> GROUP_READ NUMBER ( 1 ) NOT NULL, >>> GROUP_WRITE NUMBER ( 1 ) NOT NULL, >>> OTHER_READ NUMBER ( 1 ) NOT NULL, >>> OTHER_WRITE NUMBER ( 1 ) NOT NULL, >>> ROW_USER_ID NUMBER ( 12 ) NOT NULL, >>> ROW_GROUP_ID NUMBER ( 12 ) NOT NULL, >>> ROW_PROJECT_ID NUMBER ( 12 ) NOT NULL, >>> ROW_ALG_INVOCATION_ID NUMBER ( 12 ) NOT NULL, >>> CONSTRAINT PK_BIBLIOGRAPHICREFERNECE PRIMARY KEY >>>(BIBLIOGRAPHIC_REFERENCE_ID) >>>); >>> >>> >>> >>> >> > |
|
From: Valerie W. <va...@sa...> - 2002-07-31 14:21:32
|
Angel Pizarro wrote: > > Arnaud, > this is a bit over powered for what we use, but I think it not all > that much more, except the bit about the authors. Can I make the > compromise to keep the author field in the table in addition to a > normalized set of tables for those that choose to use them? > > ALso I had forgotten to add an (external_db_release_id, source_id) tuple > to the table before I sent it out. This would only take care of one > external reference, though. A possible solution would be to use a many to > many relation to DbRef for all of the external ids such as pubmed and > medline. > > Lastly, what is your intent with the 'is_curated' bit column? > Angel We need to be able to add a flag to distinguish between papers which have been read and manually curated, and those which have been mined but not curator reviewed to extract data Val -- ---------------------------------------------------------------------------------- Valerie Wood Tel: 01223 494954 S. Pombe Genome Project Fax: 01223 494919 The Sanger Centre email: va...@sa... Wellcome Trust Genome Campus http://www.sanger.ac.uk/Projects/S_pombe Cambridge CB10 1SA |
|
From: Angel P. <an...@sn...> - 2002-07-31 15:35:28
|
If this is from an internal project to PSU, I will create an external table for BibRef annotations, such as 'manually reviewed', 'mined', etc. I don't want to add information not intrinsic to the entry in the same table. Angel On Wed, 31 Jul 2002, Valerie Wood wrote: > > Lastly, what is your intent with the 'is_curated' bit column? > > Angel > > > We need to be able to add a flag to distinguish between papers which > have been read and manually curated, > and those which have been mined but not curator reviewed to extract data > > Val > > > |
|
From: Valerie W. <va...@sa...> - 2002-08-01 11:53:10
|
sounds fine. Angel Pizarro wrote: > > If this is from an internal project to PSU, I will create an external > table for BibRef annotations, such as 'manually reviewed', 'mined', etc. I > don't want to add information not intrinsic to the entry in the same > table. > > Angel > > On Wed, 31 Jul 2002, Valerie Wood wrote: > > > Lastly, what is your intent with the 'is_curated' bit column? > > > Angel > > > > > > We need to be able to add a flag to distinguish between papers which > > have been read and manually curated, > > and those which have been mined but not curator reviewed to extract data > > > > Val > > > > > > -- ---------------------------------------------------------------------------------- Valerie Wood Tel: 01223 494954 S. Pombe Genome Project Fax: 01223 494919 The Sanger Centre email: va...@sa... Wellcome Trust Genome Campus http://www.sanger.ac.uk/Projects/S_pombe Cambridge CB10 1SA |
×
Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.
Thanks for helping keep SourceForge clean.
X