Query Genbank Using A Mysql Interface

Question

[arronslacey]

Hi - how can I query genbank using a mysql type interface, so that I can bring back sequences of interest based on a sub-string of the Definition field? or what is the least painful way to download all proteins in genbank and put them into a mysql file? even just download the flat file database and I can work out the rest from there.

thanks

Answer 1

[Pierre Lindenbaum]

As far as I remember, NCBI uses Sybase(?) ( http://www.ncbi.nlm.nih.gov/books/NBK21081/ outdated ? ) : you cannot query it using a mysql client

. Retrieval of full documents comes from a variety of “back-end” databases, depending on the node. These might be Sybase or Microsoft SQL Server relational databases of a variety of schemas or text files of various formats. Links are supported using the Sybase IQ database product.

You should use the E-Utilities: http://www.ncbi.nlm.nih.gov/books/NBK25500/

or what is the least painful way to download all proteins in genbank and put them into a mysql file

ftp://ftp.ncbi.nih.gov/genbank/

[wdiwdi]

NCBI is Microsoft SQL Server.

[arronslacey]

thanks pierre this is great.