Thanks bob!
>> "As someone who loads data from several labs and institutes around the
>> world, I really doubt you can just point your software at a bunch of
>> directories full of raw data files and expect to have all the data loaded in
>> a meaningful way. It does unfortunately take some curation. Perhaps with
>> Affy data you have less to worry about (ready-annotated probesets, only one
>> channel, no dye swaps etc), but maybe you don't just have Affy? I doubt the
>> experimental factor annotations will just fall into place without some
>> curation."
Agreed that no such magic-mapping utility exists. Fortunately, our initial
scope is limited to 3 large labs and only Affy chips (500k SNP, CGH, and
expression arrays, respectively). As for MIAME compliance, I concur that
experimental purpose and protocols are not "gleaned" from the raw results, and
will indeed need to be read from the corresponding local systems.
-->Where do you believe the low hanging fruit is? <----
We have hundreds of annotated(?) arrays, teams of computational biologists, and
investigators that are only willing to share unpublished results IFF the
results dont go into a public repository such as GEO.
Thoughts?
--andy
McMurry, Andrew J. writes:
> Thank you Nick, Tony:
> To be clear, there are three distinct use cases:
> (1) UI for complete LIMS management system
> (2) API for reading existing microarray results and loading them into a DB
> (3) Querying that database using webservices such as those supported by
> BioMoby
>
> I am primarily interested in #2 -- you are probably not suprised to find
> that harvard medical school has troves of unpublished array results (affy
> cel files, etc) scattered across many desktops just waiting to be uploaded
> to locally controlled repository (not GEO, unpublished results). Hence, the
> Base2 API could be attractive if I dont have to rewrite the array file
> parsers for CEL/MINIML/etc. I'm hoping that user intervention wouldn't
> really be required, and that I could write an adapater for Base2 that would
> simply INSPECT a folder of say, CGH results, and load it into a local Base2
> instance on a scheduled basis. (our investigators are always over stretched
> for time and will likely not take the time to submit each sample
> individually).
>
As someone who loads data from several labs and institutes around the world, I
really doubt you can just point your software at a bunch of directories full
of raw data files and expect to have all the data loaded in a meaningful way.
It does unfortunately take some curation. Perhaps with Affy data you have
less to worry about (ready-annotated probesets, only one channel, no dye swaps
etc), but maybe you don't just have Affy? I doubt the experimental factor
annotations will just fall into place without some curation.
> As for #3, these locally controlled instances would (later) indeed need to
> be queried and aggregated. We already have a query-aggregator technology
> (SPIN) which we are extending for this purpose. That said, the serialized
> XML query payload could be made very close to BIOMoby, if appropriate. We
> are open source advocates so any upgrades/extensions would be of course
> donated back to this community freely.
>
> thoughts?
> --andy
cheers,
Bob.
--
Bob MacCallum | VectorBase Developer | Kafatos/Christophides Groups |
Division of Cell and Molecular Biology | Imperial College London |
Phone +442075941945 | Email [EMAIL PROTECTED]
-------------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
_______________________________________________
basedb-devel mailing list
basedb-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/basedb-devel
-------------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
_______________________________________________
basedb-devel mailing list
basedb-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/basedb-devel
