Found a bug in the Node Source panel

[Amy Olex]

Hello,

I just got an error message while using BioNetBuilder and it told me
to send an email to this address. I was trying to create a network
for mus musculus with my own list of node ID (Entrez GeneID). I
imported the list (16,375 nodes) into the Node Sources panel and then
clicked Next. After a few minutes this message appeared. If you can
help with that that would be great! I have Window's Vista with
Cytoscape 2.6.3 installed.

Thanks,
-Amy

--
Amy L. Olex
Bioinformatics Specialist
Department of Computer Science
Wake Forest University, Winston-Salem, NC

[Kevin Drew]

Hi,

The server ran out of heap space because of a few large queries. I
have restarted it and it should be good to go again. Please let me
know if you have anymore trouble.

Thanks,

Kevin Drew
Bonneau Lab

> --
> You received this message because you are subscribed to the Google
> Groups "BioNetBuilder" group.
> To post to this group, send email to bionet...@googlegroups.com.
> To unsubscribe from this group, send email to bionetbuilder
> +unsub...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/
> group/bionetbuilder?hl=en.
>

[Amy Olex]

Hi Kevin,

Thank you for restarting the server. It is working now :) I have a
question for you though. I have a list of about 16,000 entrez IDs,
and I want to get all the KEGG mouse interactions for. I submit the
list of nodes and calculate the number of edges in the KEGG database
(just over 47,000 was the number) and then I ask it to create the
network. After the BioNetBuilder bar disappears I have a network, but
it says it has zero nodes and zero edges. I right click and say
'CreateView' and the nodes and edges then appear along with the
network view. However, there were only about 2,000 nodes and 29,000
edges that appeared instead of the 16K and 49K that the wizard showed.
Plus the edges were not colored like the tutorial said. Do you know
why this is? I thought maybe the KEGG edge number just gave me all
edges instead of only Mouse, but when I input a smaller network (about
6,000 nodes) it listed fewer edges then only imported fewer still for
the final network. Do you know why this is? Do I need to run a
different version of Cytoscape that is more compatible?

Thanks!
-Amy

>> bionetbuilde...@googlegroups.com.

>> For more options, visit this group at
>> http://groups.google.com/group/bionetbuilder?hl=en.
>>
>
>

--

[Kevin Drew]

Hi,

I think the problem has to do with the translation of entrez IDs.
The node source panel will read in the list of 16k ids and report how
many were read but those need to be translated to another identifier
(gi) to make it compatible with the other databases. This
translation may not have worked for all of the ids in the list and
therefore produce a network with a smaller number of nodes. You
might want to try to use gi which may give you better results.

For the colors, you can try setting the colors in cytoscape's "viz
mapper". I think the tutorial has instructions but they may be for
an older version of cytoscape.

Also, is there anyway to break your network into smaller networks?
Building a network this large is very memory intense and our server
is not equipped to handle such large queries and occasionally results
in the server crashing.

Thanks,

Kevin Drew
Bonneau Lab

> To unsubscribe from this group, send email to bionetbuilder
> +unsub...@googlegroups.com.

[Amy Olex]

Kevin,

Thanks for your help. I'm going to try using the GI numbers now.
Unfortunantly, I think I may have crashed your server again before I
read your email because it is not working now :( SOOO SORRY! Maybe
you can help figure out the best way to do what I need to do.

I have all the genes on the Affy Mouse GeneChip that I found GI
numbers for (over 45,000 ids). I am trying to create a global KEGG
network for all these IDs for the Mouse Genome using KEGG. Since you
said 16K ids were a lot for your system, I'm thinking 45K will be way
to many. If I break these into lists of about 4800 to 5000 ids at a
time would your server be able to handle those? Also, if I do that
will BioNetBuilder be able to merge the networks and add in the
missing edges?

Thanks, and I will really try not to crash the server again. I think
what did it was I tried leaving the Node Selection tab blank and it
returned over 400K edges. When I tried to create the network it
fizzled out on me then wouldn't work anymore. Again...sooo sorry.

Thanks for all your help!
-Amy