@ Pierrie : I know gene number varies between the three databases. Is there any significant differences between the SNP recorded in NCBI, UCSC and Ensembl.
I only asked once the dbSNP helpdesk about archiving, regarding information mapped to previous human genome builds, and what they told me at that time (it was on 2008) was that they were only able to maintain I single human genome build on their systems, and that if I needed older versions I would have to query them through the UCSC browser. as you may imagine, this was not very helpful.
I know this is not exactly answering your question, and that there is already an answer citing Ensembl's Biomart, but IMHO that's the best site to place such queries. in fact, there are 3 main reasons why I would highly suggest you to consider working with Ensembl to retrieve this kind of data:
they DO maintain an archive of all
previous builds, so you may always
be able to trace all your work, at
any time in the future
their API not
only is robust, but also it is very
well documented
they keep track of
all "external" information they
mirror, so even if they assign their
own names to genes and variants, you
are always able to find your NCBI
coded information
it's not that I am an Ensembl fan against NCBI, but I've found out that when you need to dig really deep into the querying system you find more answers on the european site rather than on the american. but as you are a very experienced user, I guess you will already have another opinion. I just wanted to point this out, in case it helps any other reader coming to this thread looking for advice ;)
I'm reasonably sure that EUtils doesn't handle segment queries, just because I've never seen any documentation or code examples. Either BioMart or UCSC would be the "standard" way to approach this task. Their relational database structure makes it very easy.
I'd guess that it's possible to hack together a solution; perhaps based around the Map Viewer data, or by joining datasets from multiple EUtils queries, but it might not be the most efficient approach.
Yes, thanks Pedro. I can also use BioDas:UCSC, Biodas:Ensembl, mysql etc... But I wish I could get a tool integrated with the NCBI data.
@ Pierrie : I know gene number varies between the three databases. Is there any significant differences between the SNP recorded in NCBI, UCSC and Ensembl.
Pedro Paulo, all the same thing :-). Anyway, I'm not aware of a way to do with NCBI tools, maybe something in the toolkit might help then.