Re: [cytoscape-helpdesk] overlay gene expression data onto networks!

[piet molenaar]

Hi, It sounds as though you did everything right. Are you sure the values are actually mapped to the nodes? You can check this in the datapanel by selecting show all attributes and select all nodes in your network.

Piet

On Fri, Jan 11, 2013 at 6:46 PM, PS <padmapriya...@gmail.com> wrote:

Hi All,

I am in the process of overlaying gene expression data(of Oats) onto rice pathways using Cytoscape. I have kegg rice pathway files in kgml format. It has enzyme commission numbers in the network file.  I have a tab delimited text file prepared from excel which has two columns Enzymeno and Foldchange.

I am able to load the network in cytoscape without any problem.

 When I am importing gene expression data onto the network (File --> Import --> Attribute/Expression matrix) I select the file in the dialog box and there is another place that says 'Assign values to nodes using'  ...(default it shows ID). I left it as "ID" .

 After I import the file, I go to Vizmapper tab and then in Visual mapping browser I set the "Node Color" to  Foldchangeexp which is fold change ratio. This is the attribute that I would like to link to node color. I selected continuous mapper and set the color for the different values in the Gradient Editor for Node Color.  After clicking on OK I do not see that there is any change in the network node colors.

I am not sure if there is anything that I missed to do because I do not see any color change in the network. Will I be able to see any color change in the enzyme commission number in the pathways?

Finally I like to see how the enzyme commission number boxes in the networks are changing colors due to different fold change values.

Thanks for all your suggestions in advance.

PS

--
You received this message because you are subscribed to the Google Groups "cytoscape-helpdesk" group.
To view this discussion on the web visit https://groups.google.com/d/msg/cytoscape-helpdesk/-/TnW5kfTEMS0J.
To post to this group, send email to cytoscape...@googlegroups.com.
To unsubscribe from this group, send email to cytoscape-helpd...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/cytoscape-helpdesk?hl=en.

--
Piet Molenaar
piet...@gmail.com
Department of Oncogenomics, M1-131
Academic Medical Center
University of Amsterdam
Meibergdreef 9
1105 AZ Amsterdam
the Netherlands

tel (+31) 20-5666592
fax (+31) 20-6918626

[PS]

Hi Piet,

Thanks for your response.

I checked in the data panel by selecting all attributes and all nodes in the network but the values of all the columns show except fold change ratio which is the gene expression column does not show any values and this column is empty.

When I loaded my gene expression data which is a text tab delimited file with extension of .pvals or .txt both seem to give the same problem. As soon as I loaded the gene expression data file it says in a small window:

Loading Gene Expression data
Data read from ecfoldchange-jan8.pvals.txt
No of genes 5257
No of conditions 1
significance values no
Minimum expression -999999.99
Maximum expression +999999.99

All the above details are correct. There are total of 5257 entries from my text file(2 columns Enzymeno and Foldchange). Each of those entries has Enzymeno and Fold change ratio. Still when showing in the data panel Foldchange column is empty.

I am not sure why the data cannot map to nodes(enzymecommission numbers) in the networks.

Thanks for your help.

PS

On Monday, January 14, 2013 3:53:46 AM UTC-6, Piet wrote:

Hi, It sounds as though you did everything right. Are you sure the values are actually mapped to the nodes? You can check this in the datapanel by selecting show all attributes and select all nodes in your network.

Piet

On Fri, Jan 11, 2013 at 6:46 PM, PS <padmapriya...@gmail.com> wrote:

Hi All,

I am in the process of overlaying gene expression data(of Oats) onto rice pathways using Cytoscape. I have kegg rice pathway files in kgml format. It has enzyme commission numbers in the network file.  I have a tab delimited text file prepared from excel which has two columns Enzymeno and Foldchange.

I am able to load the network in cytoscape without any problem.

 When I am importing gene expression data onto the network (File --> Import --> Attribute/Expression matrix) I select the file in the dialog box and there is another place that says 'Assign values to nodes using'  ...(default it shows ID). I left it as "ID" .

 After I import the file, I go to Vizmapper tab and then in Visual mapping browser I set the "Node Color" to  Foldchangeexp which is fold change ratio. This is the attribute that I would like to link to node color. I selected continuous mapper and set the color for the different values in the Gradient Editor for Node Color.  After clicking on OK I do not see that there is any change in the network node colors.

I am not sure if there is anything that I missed to do because I do not see any color change in the network. Will I be able to see any color change in the enzyme commission number in the pathways?

Finally I like to see how the enzyme commission number boxes in the networks are changing colors due to different fold change values.

Thanks for all your suggestions in advance.

PS

--
You received this message because you are subscribed to the Google Groups "cytoscape-helpdesk" group.
To view this discussion on the web visit https://groups.google.com/d/msg/cytoscape-helpdesk/-/TnW5kfTEMS0J.
To post to this group, send email to cytoscape...@googlegroups.com.

To unsubscribe from this group, send email to cytoscape-helpdesk+unsub...@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/cytoscape-helpdesk?hl=en.

[piet molenaar]

Hi,

Apparently the values are not properly mapped; can you send me a sample of your data?

Cheers,

Piet

To view this discussion on the web visit https://groups.google.com/d/msg/cytoscape-helpdesk/-/BAjFMCDLQqAJ.

To post to this group, send email to cytoscape...@googlegroups.com.

To unsubscribe from this group, send email to cytoscape-helpd...@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/cytoscape-helpdesk?hl=en.

[PS]

Hi Piet,

Thanks for offering to look at the data.

Please refer the attached text file(first50.pvals) for the first 50 entries which has two columns ID and Foldchange. Initially I had the first column named as Enzymeno. I renamed Enzymeno column as ID and checked again to see if the column name ID would help but it didn't make a difference.

I appreciate your help.

PS

To unsubscribe from this group, send email to cytoscape-helpdesk+unsubscribe@googlegroups.com.

[PS]

Hi Piet,

Information about pathways that I overlay the gene expression data:

1. Rice kegg pathways(This has enzyme commission numbers in the network after it is imported inside Cytoscape) in xml format. These KGML files could be read by kgmlreader plugin in cytoscape.

Thanks again.

PS

[piet molenaar]

Hi,

I think you'll want to use the File>Import> Attribute from Table(txt/Msexcel) functionality. In the dialog that pops up you have to select the mapping options checkbox, this allows you to select the proper id column in the datafile you provided to map to the proper (Enzyme id ) in the network. It is described here in the documentation: http://cytoscape.org/manual/Cytoscape2_8Manual.html#Import Attribute Table Files

I tried one of the rice kegg xml files but these did not contain EC id's to map your data on.

If you continue to have problems, you can send me one of the kegg files containing EC's also

Cheers,

Piet

To view this discussion on the web visit https://groups.google.com/d/msg/cytoscape-helpdesk/-/zbHjk_z_4TUJ.

To post to this group, send email to cytoscape...@googlegroups.com.

To unsubscribe from this group, send email to cytoscape-helpd...@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/cytoscape-helpdesk?hl=en.

[PS]

Hi Piet,

Thanks for sending the link for importing node attributes.

I did try to do File ->Import Attribute from table using a text file with .pvals extension. It shows the two columns (ID and Foldchange) when I try to import with tab as delimiter. It shows there are 5257 entries.

I am still not very sure of which attribute in the network should I map to the ID column in my gene expression data file. There are key attributes in the network like ID, KEGG entry and canonical name that shows in the Import dialog box. If I say ID to ID doesn't work.  Moreover I see in the data panel that no data is showing up.

I have attached the rice kegg pathway file (kgml file) in xml format.

I will keep trying and post the updates to you.

I appreciate your time.

PS

On Tuesday, January 15, 2013 3:17:04 AM UTC-6, Piet wrote:

Hi,

[piet molenaar]

Hi,

Sorry for the delay, but find attached a session file where I mapped the kegg.label attribute to the ID in the pvals.txt file. In the vizmapper I mapped the color to the values of the foldchange. You might have to adapt some of the id values that contain '-'

Hope this helps as an example

Piet

--
You received this message because you are subscribed to the Google Groups "cytoscape-helpdesk" group.

To view this discussion on the web visit https://groups.google.com/d/msg/cytoscape-helpdesk/-/eG0edkRw-QwJ.

To post to this group, send email to cytoscape...@googlegroups.com.
To unsubscribe from this group, send email to cytoscape-helpd...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/cytoscape-helpdesk?hl=en.

[PS]

Hi Piet,

Thanks for the details about mapping the network attribute with the expression data. I tried and I am glad it works now.

Now  I will have to find a way to map same Enzyme commission numbers but different fold change ratios because some of my entries will have two or three fold change values for the same ecno. So my first entry of enzyme commission number will be mapped to the first entry of fold change but not the other one or two fold change values.

say 1.2.3.1   with fold change of -5.5
again 1.2.3.1 with fold change of 76.5

Regarding the enzyme commission numbers which has - in the entries, we are not certain about the fourth part of the ecno(say 1.3.1.-)

 I appreciate your help in finding the solution for the problem and all your responses.

Thanks again.

PS

To unsubscribe from this group, send email to cytoscape-helpdesk+unsub...@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/cytoscape-helpdesk?hl=en.

[piet molenaar]

Hi,

You can load multiple attributes to the same id in one go; you just have to name them differently; otherwise they'll be overwritten. You cannot map multiple values to the same color, if that is what I  understood ;-)

But there are plugins that allow you to step through the different sample points in your expression dataset; look into the app store (http://apps.cytoscape.org/) for e.g. vistaclara, or dynamic expr plugins.

Hope this helps,

Piet

To view this discussion on the web visit https://groups.google.com/d/msg/cytoscape-helpdesk/-/3t8u7v1WxzUJ.

To post to this group, send email to cytoscape...@googlegroups.com.

To unsubscribe from this group, send email to cytoscape-helpd...@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/cytoscape-helpdesk?hl=en.

[Scooter Morris]

On 1/18/13 1:02 AM, piet molenaar wrote:

Hi,

You can load multiple attributes to the same id in one go; you just have to name them differently; otherwise they'll be overwritten. You cannot map multiple values to the same color, if that is what I  understood ;-)

But there are plugins that allow you to step through the different sample points in your expression dataset; look into the app store (http://apps.cytoscape.org/) for e.g. vistaclara, or dynamic expr plugins.

Hope this helps,

Piet

clusterMaker will also do this, and do the standard hierarchical (or k-means) clustering.

-- scooter

--
You received this message because you are subscribed to the Google Groups "cytoscape-helpdesk" group.

[PS]

Hi Piet,

 

Thanks for the info about the plugins. I will look into the details and see which plugin which help in visualizing different gene expression values at the same time for the same enzyme commission number.

 

PS

To unsubscribe from this group, send email to cytoscape-helpdesk+unsubscribe@googlegroups.com.

[PS]

Hi Scooter,

 

Thanks for the suggestion. I will read about clustermaker and get back if I have questions.

 

PS

To unsubscribe from this group, send email to cytoscape-helpdesk+unsub...@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/cytoscape-helpdesk?hl=en.

--
Piet Molenaar
piet...@gmail.com

Department of Oncogenomics, M1-131
Academic Medical Center
University of Amsterdam
Meibergdreef 9
1105 AZ Amsterdam
the Netherlands

tel (+31) 20-5666592
fax (+31) 20-6918626

--

You received this message because you are subscribed to the Google Groups "cytoscape-helpdesk" group.

To view this discussion on the web visit https://groups.google.com/d/msg/cytoscape-helpdesk/-/3t8u7v1WxzUJ.

To post to this group, send email to cytoscape...@googlegroups.com.

To unsubscribe from this group, send email to cytoscape-helpdesk+unsub...@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/cytoscape-helpdesk?hl=en.

--
Piet Molenaar
piet...@gmail.com
Department of Oncogenomics, M1-131
Academic Medical Center
University of Amsterdam
Meibergdreef 9
1105 AZ Amsterdam
the Netherlands

tel (+31) 20-5666592
fax (+31) 20-6918626

--
You received this message because you are subscribed to the Google Groups "cytoscape-helpdesk" group.
To post to this group, send email to cytoscape...@googlegroups.com.

To unsubscribe from this group, send email to cytoscape-helpdesk+unsub...@googlegroups.com.