[BioSQL-l] The taxon table

[Yves Bastide]

Greg Tyrelle wrote:
> *** Yves Bastide wrote: 
>   |Er, I said postgresql specific :)
> 
> Sorry, I meant porting it to mysql :)

Thought so.  Postgresql's COPY FROM format is:
* one tuple per line
* tab-delimited fields
* NULL is \N

> 
>   |Put at <http://www.irisa.fr/symbiose/people/bastide/load_tax.py> anyway; 
>   |you're welcome to use it.
> 
> Thanks.
> 
>   |>Also is taxonomy optional in BioSQL or will any fresh install of
>   |>BioSQL need to load the ncbi taxonomy data by default ?
>   |
>   |The taxonomy is optional; furthermore, species can be added at will 
>   |using e.g. biosql-pg's persistence.
> 
> Cool, I can't say I understand postgresql persistence but I'll check
> it out.

(Oops, of course I meant bioperl-db...)

> 
> Currently the BioSQL modules in biopython seem to add new taxa 
> id from genbank records, I was wondering how the new biosql taxon
> tables will effect this if the ncbi taxonomy database is not preloaded ?
> 
> Does that make sense or should I be asking this on the biopython lists ?
> 

biopython's BioSQL first lookup by binomial, then by NCBI taxon id, and 
inserts a new record if needed.

However, biopython's BioSQL is currently using the pre-Singapore data 
model.  I'll update it soon if no one beats me to it <.4 hint>

> _greg
> 

yves