protein interaction network on circos

261 views
Skip to first unread message

teragene

unread,
May 20, 2009, 8:21:43 PM5/20/09
to Circos
Hi all

I want to draw protein interaction relationships on circos.
Do you have an idea how to deal with karyotype definition?
I mean is it possible plot (draw) 'PROTEIN PROTEIN INTERACTION
NETWORK' with circos?

cheers

Martin

unread,
May 20, 2009, 8:27:40 PM5/20/09
to Circos

Yes, it's possible, as long as you can represent your data as pairs of
entities.

You can represent each protein in the karyotype file as if it were a
chromosome. You can set the length to be the size of the protein, or
make each the same size.

chr - protein001 001 0 100 black
chr - protein002 002 0 200 black
...

Here I made protein001 100 have size 100, and protein002 have size
200.

In the link file, you can create links between them like this

link_001 protein001 50 50 # middle of protein object, or use 0 100 to
set link to span the whole protein
link_001 protein002 100 100 # same as above - middle of protein002

Shiva

unread,
May 31, 2012, 7:14:26 AM5/31/12
to circos-data-...@googlegroups.com
Hi Martin,

Thanks for creating this wonderful software. I am now using this to create genetic interactions between the genes. But, I could see many false links shown between the genes (for example between TSC1 and KDM5C gene). My link input file actually do not contain any links for these genes. Let me know how to rectify this ?

Anticipating your reply

Thanks!
Shiva
geneticInteractions_links_genes.txt

Martin Krzywinski

unread,
Jun 5, 2012, 6:39:38 PM6/5/12
to circos-data-...@googlegroups.com

Links are defined between two positions. 

In old versions of Circos, the format was 

LINKID chr1 start1 end1
LINKID chr2 start2 end2

and in newer versions, a link is specified on one line

chr1 start1 end1 chr2 start2 end2

Your file contains fields like

ACVR1 hs1 253 254
ACVR1 hs4 13 14
ACVR1 hs4 5 6
AFF2 hs1 167 168
AFF2 hs1 233 234
AFF2 hs1 27 28
AFF2 hs1 57 58
...

which is parsed as the old format. The trouble with it is that you're defining a link with more than two lines. 

To represent a link between geneA at chr1:10-20 and geneB at chr2:20-30, you would use

chr1 10 20 chr2 20 30

You can assign the link an id to keep track of the gene names involved

chr1 10 20 chr2 20 30 id=geneA_geneB

This would allow you to later use rules to format the links based on gene names.

Martin Krzywinski
science + art




--
You received this message because you are subscribed to the Google Groups "Circos" group.
To view this discussion on the web visit https://groups.google.com/d/msg/circos-data-visualization/-/pkUkaMZ5W2IJ.
To post to this group, send email to circos-data-...@googlegroups.com.
To unsubscribe from this group, send email to circos-data-visual...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/circos-data-visualization?hl=en.

Reply all
Reply to author
Forward
0 new messages