Re: [Heliconiusdb-devel] more questions about the data in the mcmillan.psql file downloaded from the SourceForge site

[Hilmar L. <hl...@du...>]

Hi Xianhua -

good questions, and I hope Owen can provide some insight here. Note  
that if at all possible we should try not to use the .psql file - if  
I'm not mistaken these are from an intermediary (between the original  
and the eventual target) database.

If at all possible we should try to write import scripts that take  
the FileMaker exports, so that we have a documented data migration  
path from the original database to the new format. There were rumors  
that the FileMaker exports contain some inconsistencies that need to  
be fixed - I would like these to be documented, and if possible  
implemented in code, or alternatively the records that need special  
and manual treatment need to be separated out into their own file.

	-hilmar

On May 7, 2007, at 12:58 PM, Xianhua Liu wrote:

> Hello:
>
> I am trying to import McMillan's data into the new heliconius  
> database. Here are the questions about the data in the  
> mcmillan.psql file downloaded from the SourceForge site:
>
> 1. There are 10 tables:
>
> biotype
> biotype_organism
> individual
> m_overall
> m_reared
> m_wild
> organism
> overall_reared
> overall_wild
> pedigree
>
> Among these tables, the structure and meaning of biotype,  
> biotype_organism and organism seem clear to me. But other tables  
> seem to contain data with overlaps. I am not sure which tables I  
> should use for non-hybrids, natural hybrids and broods data.
>
> 2. In several tables, there are columns called something old, such  
> as "r_old_parent" in "m_reared", and something calculated, such  
> "r_calc_mother" and "r_calc_father". Which columns should be used?
>
> 3. There are "*genus", "*species" and "*race" columns in the  
> several tables. What does "race" mean? subspecies? For example,   
> there is a record in the "overall_reared" table with "Heliconius"  
> as genus, "hybrid erato x himera" as species, and "petiverana x  
> himera" as race. How to decomposite these names into a biotype  
> record associated with multiple organisms?
>
> 4. There are "r_old_cross_type", "r_calc_cross_type" and  
> "r_calc_filial_status" in the "overall_reared" and "m_reared"  
> tables. Which should be used to populate the type of a  
> crossexperiment?
>
>
> Thanks,
>
> Xianhua
>
> ---------------------------------------------------------------------- 
> ---
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/ 
> _______________________________________________
> Heliconiusdb-devel mailing list
> Hel...@li...
> https://lists.sourceforge.net/lists/listinfo/heliconiusdb-devel

-- 
===========================================================
: Hilmar Lapp  -:-  Durham, NC  -:- hlapp at duke dot edu :
===========================================================