[Bioperl-l] RE: Bioperl DB successful queries

[Law, Annie]

Hi Hilmar and other bioperlers,

Thanks for your response.  I would appreciate help with the following.

1.  If my starting point is Unigene Cluster id.  These are the select
statements that I am using to Get to locuslink id.  I think it could be
cleaned up a bit but it seems to be doing the job. It seems that it was
indicated that these steps were incorrect.  I just wanted to be more
explicit since it seems to be correct.
Here are the SQL statements I use in the following order.
$sthdbx = $dbh->prepare("Select dbxref_id from dbxref where accession =
'$unigeneans_curr' and dbname = 'Unigene'");

$dbxref_loc_ans[0] is answer from first select statement.
$sthbio = $dbh->prepare("Select bioentry_id from bioentry_dbxref where
dbxref_id = $dbxref_loc_ans[0]");
$sthloc = $dbh->prepare("Select accession from bioentry where bioentry_id =
$bioentry_loc_ans[0]");
Answer of this last select would be the locuslink id

2. I'm not sure how Subject Bioentry in Bioentry Relationship 
is the Unigene Cluster ID.
For example, I execute the following SQL statements in the following order
SELECT * FROM 'bioentry' where accession = 'H72976'  ANSWER: bioentry_id
795253 SELECT * FROM 'bioentry_relationship' where object_bioentry_id =
'795253'  ANSWER: subject_bioentry_id 795131 I plugged my clone id into
SOURCE and got a unigene id Hs. 39488 which does not seem to match the
subject_bioentry_id 795131 Am I missing a step?

3.  The script load_seqdatabase.pl is flexible and useful.  If one is not
concerned about speed.  I would like to know what
The disadvantage and advantage is of emptying the bioperl-db tables and then
reloading information  from scratch vs. loading
A full database with mergeobjs and lookup options?

Thanks very much,
Annie.