Phenote .ace citace upload

From WormBaseWiki
Jump to: navigation, search

back

Dumping phenotype app_files into .ace

Two dumping scripts are required to capture all of the allele phenotype data, both on tazendra:

/home/acedb/karen/WS_upload_scripts/phenotype/use_package.pl*
/home/acedb/karen/WS_upload_scripts/paper_object/get_paper_object.pl*

allele_phenotype and paper_object are dumped on a weekly cron job Sundays at 4am
see /home/acedb/cron/crontab_20140703

0 4 * * sun /home/acedb/karen/WS_upload_scripts/paper_object/get_paper_object.pl
0 4 * * sun cd /home/acedb/karen/WS_upload_scripts/transgene; ./use_package.pl
0 4 * * sun cd /home/acedb/work/allele_phenotype/; ./use_package.pl


/home/acedb/karen/WS_upload_scripts/phenotype/use_package.pl : This script creates 5 files:

  • allele_phenotype.ace.<timestamp>
  • var_phen.ace
  • mol_phene.ace.<timestamp>
  • mol_phene.ace
  • err.out.<timestamp>

var_phen.ace and mol_phen.ace are copies of allele_phenotype.ace and mol_phene.ace.
var_phen.ace and mol_phen.ace are picked up by spica on a cron job that is run every Monday at 8am.

/home/acedb/karen/WS_upload_scripts/paper_object/get_paper_object.pl*:

This script creates a file that will associate phenotype annotated alleles with their respective genes.
This is needed so references from phenotype annotations are added to the references on the respective gene page. And example from this file:

Paper : WBPaper00000005
Gene	WBGene00000898
Allele	"WBVar00143949"	Inferred_Automatically	"Inferred automatically from curated phenotype"
Gene	WBGene00004015
Allele	"WBVar00000103"	Inferred_Automatically	"Inferred automatically from curated phenotype"
Gene	WBGene00004015
Allele	"WBVar00000103"	Inferred_Automatically	"Inferred automatically from curated phenotype"

Dumping allele_paper connections into .ace file

is at :
/home/acedb/..../get_paper_object.pl

Run it by going to that directory and redirecting output to a file

./get_paper_object.pl > file 

If there are errors, which is rare, they will look something like:

// ERR 
There's an object that has no name and no type, but still has a paper.

It has postgres database ID 2712 testdb=> SELECT * FROM app_tempname WHERE joinkey = '2712'; joinkey | app_tempname  
 |app_timestamp 
----
+--------------+---------------
...
----
+------------------+---------------
...
----
+-----------------+-------------------------------
2712 | WBPaper00002087 | 2006-07-11 11:52:20.208295-07 (1 row) Here's all the data for it when it got migrated 
from the  CGI  

Testing Phenote .ace for upload

Evaluate how the uploaded data impacts WB by comparing object numbers before and after data upload. This means that you need access to the latest WS and have the means to edit the WS to read the .ace dump.

Test file locally on Acedb

This step is to test two things: First, that the file is readable. Second, that the file does not alter any objects it shouldn’t.

Launch local acedb:

$ cd Desktop/acedb
$ ./xace /.../WSXXX/acedb

Comparing data builds: From the latest build, record

  1. Number strain objects
  2. Number life stage objects
  3. Number anatomy terms
  4. Number variation'-phenotype connection'
use either AQL query:
select all class variation where exists_tag ->phenotype select p, p-> variation from p in class 
phenotype where exists_tag  p->variation 
5. Number variations that are alleles:

use WQL

find variation variation_type=allele

or AQL

select a from a in class variation where exists_tag a->allele 
6. Number alleles with a phenotype

Find all alleles with a phenotype

select v from v in class variation where exists_tag v->phenotype

Check that the .ace dump file is readable:

to load in .ace dump

Select ‘Edit..‘ ‘Read .ace file’,  Accept change in write privileges,  Select ‘Open ace file’,  Find and select file to open,  Select ‘Read all’

If the .ace file is okay, then 100% of the lines will have been read in. This will be noted in the second line called ‘Line:’. If the dump produced a bad file, the read in will stop at the point where the problem occurred. You can go to that line in the .ace file and check it out.

Check that the file does not alter any objects it shouldn’t:

Once .ace dump is loaded in to acedb,  redo counts for all objects as above. Compare object numbers between latest database and new .ace file to make sure data in the dump looks reasonable, i.e., no lost data or inflated numbers.

Uploading .ace for Wen into Citace

Deposit .ace for Wen in citace:

scp var_phen.ace xxxx@xxxx.caltech.edu:/.../Data