Thanks Terry! That is exactly what I wanted, and it ran perfectly in IRIS!  I'll work on wrapping my mind around how to teach people how to create queries like that themselves.  The simplest thing that comes to mind is something like StackOverflow where people post little questions (like mine) and experts put in how to do it correctly.

Ben

--------------
Email: BPBowen@lbl.gov
Phone: (510) 214-6683
MS 977-221, Lawrence Berkeley Lab
Web: ms.lbl.gov


On Wed, Apr 10, 2013 at 12:47 PM, Terry Disz <disz@mcs.anl.gov> wrote:
Ben,

Ross and I took this up and we produced this example - I think it does what you want.
We went to the ERDB browser and followed the path from Compound->ParticipatesAs->LocalizedCompund->IsInvolvedIn



all_entities_Compound -a | grep -i valine | get_relationship_ParticipatesAs -c 1 -to id | get_relationship_IsInvolvedIn -to id,name

This produces information for all compounds with valine in the name.

Here are the first couple lines of output

disz@ash:~$ all_entities_Compound -a | grep -i valine | get_relationship_ParticipatesAs -c 1 -to id | get_relationship_IsInvolvedIn -to id,name
kb|cpd.153 L-Valine val-L cpd00156 0 1340988351 117 C5H11NO2 0 -86.49 0.43806 kb|loc.0.c.1376 kb|rxn.99 L-valine transport in via proton symport
kb|cpd.153 L-Valine val-L cpd00156 0 1340988351 117 C5H11NO2 0 -86.49 0.43806 kb|loc.0.c.1376 kb|rxn.485 Branched chain amino acid:H+ symporter (Valine)

Is this what you were looking for?

Terry



On Apr 10, 2013, at  2:17 PM, Ben Bowen <ben.bowen@gmail.com> wrote:

I just made a small update to the doc, but probably didn't make it clear.  The thought I'm trying to convey is that we are missing out on a huge opportunity by not having the API commands that perform ER-queries easier to use.  To me, the ER-queries make the KBase platform very distinct (and more powerful) from other systems.  I just can never seem to get them to work... maybe its just me, but somehow I doubt it.  Today, I was trying to get a list of reactions that Valine was involved in and I gave up after 30 minutes.  Actually, what I was trying to do was this:
1) get all the relationships that connect "Compounds"
2) for Valine, get all the results of those relationships (ie: Media IDs, Biomass terms, Reactions, etc).
What a powerful query! and what a success it would be to somehow make it easier.  I don't know what this would take, but I'm happy to talk with whoever wants to try and figure out what it would take.

Ben


--------------
Email: BPBowen@lbl.gov
Phone: (510) 214-6683
MS 977-221, Lawrence Berkeley Lab
Web: ms.lbl.gov


On Tue, Apr 9, 2013 at 12:18 PM, Paramvir Dehal <psdehal@lbl.gov> wrote:
Adam want's our Platform updates by Thursday, please update the google doc:


Let's populate this as much as we can before trying to prioritize. 

_______________________________________________
Microbe mailing list
Microbe@lists.kbase.us
https://lists.kbase.us/mailman/listinfo/microbe


_______________________________________________
Microbe mailing list
Microbe@lists.kbase.us
https://lists.kbase.us/mailman/listinfo/microbe