[Openchemistry-developers] Sample CSV data for mongochem?

[Eric E. Monson]

Hey Marcus,

If you could post a piece of a database I could import and use, that would be great.

The first major hurdle I had for building was that (since I didn't see the openchemistry superbuild until much later) none of the dependencies were listed on the mongochem wiki. So, I would try to run CMake and then run into some error. Then I would get that dependency built and installed, and then run in to another. After separately going through that with chemkit (which itself was hard to build), molequeue and the MongoDB dev library (had some trouble finding and building that on my system, too), it felt like quite an ordeal. (And that was with VTK, Qt, MongoDB itself already on my system, and boost and eigen being easy with homebrew.)

So, you could probably get quite a ways by having more detailed descriptions of the dependencies right on the mongochem wiki, rather than forcing people to hunt around for  them. Plus, let people know that the openchemistry superbuild is available. Plus, the issues we've been talking about with getting data in, which should include some documentation of the python scripts and pointers to getting chemkit built with python bindings since it relies on that.

On my system the File->Add New Data and File->Server Settings don't seem to do anything. Plus, Quit isn't really tied to app shutdown (exit) – to quit I have to click the red button to dismiss the main window.

Thanks for keeping in touch even though you're overseas,
-Eric


On Feb 6, 2013, at 4:32 AM, "Marcus D. Hanwell" <marcus.hanwell at kitware.com> wrote:

> On Tue, Feb 5, 2013 at 2:52 PM, Eric E. Monson <emonson at cs.duke.edu> wrote:
>> Hey Kyle,
>> 
>> I'm trying to get this to work, but I'm stuck trying to get data into the DB. When I do a CSV import, I only end up with the final molecule in the database, as if it's not generating new IDs as it goes, and is just overwriting the same data over and over for each molecule in the CSV file… I'll attach my CSV to see if there were any conversion problems from the SDF. I used the exact same string ("mass tpsa vabc rotatable-bonds") for the descriptor names as you suggested, so if there should have been other names listed, please tell me. (And, BTW, what is the File->Add New Data menu option supposed to do?)
>> 
> We definitely need to make it easier to try MongoChem out! There is no
> facility to turn Python wrapping on in the superbuild at this stage. I
> exported out small test database, I don't see anything in there and it
> is just a few of the PubChem molecules. I can push that somewhere, I
> think it is around 10MB bzipped and you should just be able to use the
> MongoDB import command.
> 
> I have been looking at issues such as long distance MongoDB too
> (currently using MongoChem in France with a MongoDB hosted in Clifton
> Park, NY - it is a little slow! Thanks for your notes - it looks like
> getting MongoChem built was relatively painless, and most of your
> issues were with getting Chemkit and its Python bindings to build - is
> that correct?
> 
> We will be working on making all three applications easier to get up
> and running, and demo. MongoChem presents a few special challenges of
> its own due to needing a database. We may be able to host a read-only
> demo instance of MongoDB for example, but then you hit latency issues.
> It is very easy to copy a database to a local instance too at that
> stage, or provide a few ready-made database dumps that could easily be
> imported to get you started.
> 
> Thanks,
> 
> Marcus