Using SBEAMS

From SPCTools

(Difference between revisions)
Jump to: navigation, search
Revision as of 00:54, 16 June 2009
Tfarrah (Talk | contribs)
(→Augmenting SBEAMS schema)
← Previous diff		 Revision as of 00:57, 16 June 2009
Tfarrah (Talk | contribs)
(→Filling in manageable tables)
Next diff →
Line 46: Line 46:
===Filling in manageable tables=== ===Filling in manageable tables===
 +* add INSERT queries to $SBEAMS/lib/sql/PeptideAtlas/PeptideAtlas_POPULATE.mssql or similar file in same/parallel directory
 +* execute by pasting into datastudio (though this doesn't properly record user ID)
 +* confirm at http://db.systemsbiology.net/devTF/sbeams/cgi/PeptideAtlas/ManageTable.cgi?TABLE_NAME=AT_query_option (or you can execute via this interface -- properly records user ID, but doesn't test query recorded in PeptideAtlas_POPULATE.mssql for correctness)
 +Recording queries in $SBEAMS/lib/sql/PeptideAtlas/PeptideAtlas_POPULATE.mssql (or the like) will make these execute automatically upon creation of a new PeptideAtlas.
===Changing CGI code=== ===Changing CGI code===

Revision as of 00:57, 16 June 2009

Contents

To query SBEAMS directly

Use Aqua Data Studio on mimas by copying ~tfarrah/bin/datastudio.sh to your own ~/bin and executing. Or, can install on desktop. Ask Eric D. for an account. 

hostname: mssql
port: default is fine
(Terry's account username: usual, password: TFmssql205)
click mssql in left sidebar, then server->properties to adjust username/pwd.
click button with mag. glass over table for query analyzer
Select a database to connect to.
Type in queries and press green triangle to submit.

Will need administrator privileges on your account if you want to do more than just view the data.

Augmenting SBEAMS schema: adding tables or columns

Edit two tables using Excel, in $SBEAMS/lib/conf/PeptideAtlas

  • PeptideAtlas_table_column.txt
  • PeptideAtlas_table_property.txt (add call to ManageTable.cgi at end for manageable tables)

Define new tables in $SBEAMS/lib/perl/SBEAMS/PeptideAtlas/Tables.pm in 3 places, then: 

cd $SBEAMS/lib/scripts/Core
CONFDIR="../../conf"
grep Pept batch_update_driver_tables.csh

Execute two commands that come out of grep. Then, 

cd ../../../doc/Proteomics
cd $SBEAMS/lib/scripts/Core
set dbtype=mssql
./generate_schema.pl \
--table_prop ../../conf/PeptideAtlas/PeptideAtlas_table_property.txt \
--table_col ../../conf/PeptideAtlas/PeptideAtlas_table_column.txt \
--schema_file ../../sql/PeptideAtlas/PeptideAtlas \
--module PeptideAtlas \
--destination_type $dbtype

Optional, to be sure you're doing what you want to do:

  • View the files PeptideAtlas_CREATETABLES.mssql and PeptideAtlas_CREATECONSTRAINTS.mssql to be sure they include the queries you want.
  • Edit those files to include only your new stuff. For latter file, important constraints are in upper "half". Don't need the housekeeping constraints near bottom.

If you don't edit, no huge speed issue. 

$SBEAMS/lib/scripts/Core/runsql.pl -u sbeamsadmin -s PeptideAtlas_CREATETABLES.mssql -i -delim GO # will need password
# -i = ignore errors
$SBEAMS/lib/scripts/Core/runsql.pl -u sbeamsadmin -s PeptideAtlas_CREATECONSTRAINTS.mssql -i -delim GO
# you might notice 2 "Column x is not the same data type as referencing column y" errors while loading this file; you can safely ignore them.

Document changes in appropriate file in $SBEAMS/doc/upgrade/0.24

Filling in manageable tables

Recording queries in $SBEAMS/lib/sql/PeptideAtlas/PeptideAtlas_POPULATE.mssql (or the like) will make these execute automatically upon creation of a new PeptideAtlas.

Changing CGI code

Random notes from Terry's log

load PA build: $SBEAMS/lib/scripts
 lib/conf/*   describe schema

Add column:
 -- edit table in table_columns file (or whatever)
      audit columns e.g. created_by
      can delete a record by setting record status to D
          still appears, but queries can't see it.
 -- execute generate_schema.pl  -- to help create SQL statement
generate_schema.pl -table_c PeptideAtlas_table_column.txt --table_p
PeptideAtlas_table_property.txt --schema ../../sql/PeptideAtlas/PeptideAtlas
--module PeptideAtlas --dest mssql
   == will generate files in ../../sql. these scripts never in svn
 -- write out SQL statement and ask Eric to execute
 -- document changes in $SBEAMS/doc/upgrade/0.24
       (browse through this)
 -- run:
        cd $SBEAMS/lib/script/Core
       ./update_driver_tables.pl \
         ../../conf/PeptideAtlas/PeptideAtlas_table_column.txt
       (example gleaned from $SBEAMS/doc/upgrade/0.24)

Mysql slightly diff from our sql server. Latter is preferred. but
  PeptideAtlas is by and large mysql compliant

I can monkey around in PeptideAtlas_Test with impunity
  Dave has changed this for me.

I should set my devTF (in SBEAMS.conf) to send my log info to my own
  log files

Miscellaneous useful SQL queries

If you need to change schema constraints after defining your schema:

  1. Change the constraint in $SBEAMS/conf/PeptideAtlas/PeptideAtlas_table_column.txt
  2. Actually make the schema change: 
ALTER TABLE employee  ADD CONSTRAINT pk_employee PRIMARY KEY (EmployeeId)
ALTER TABLE employee DROP CONSTRAINT pk_employee
Retrieved from "http://tools.proteomecenter.org/wiki/index.php?title=Using_SBEAMS"
Personal tools