color edges

[svenmh]

color coding (e.g. edges --- genetic: green, physical: yellow)

[svenmh]

Each edge represents more then one edge from REST. Not sure how to split them.

[svenmh]

Done

[ChristieChang]

yellow = physical, green = genetic, but when an edge represents both physical and genetic interactions, it defaults to yellow. New color for edges that are both physical and genetic? Or another way to represent this?

[RoseO]

Hope you don't mind me re-opening this, but can you please address the issue Christie raised above?

yellow = physical, green = genetic, but when an edge represents both physical and genetic interactions, it defaults to yellow. New color for edges that are both physical and genetic? Or another way to represent this?

[svenmh]

If an item is both physical and genetic it put is both yellow and green edges. I guess I can try to put a strait line between nodes.

[svenmh]

Disregard my previous message, but according to the TAB2 docs a single row can never be both.

http://wiki.thebiogrid.org/doku.php/biogrid_tab_version_2.0

column 13 says: Experimental System Type. This will be either “physical” or “genetic” as a classification of the Experimental System Name.

Unlike row 18 where items can be High Throughput, Low Throughput or Both (separated by “|”)

can you give me an example edge that is both?

[ChristieChang]

I think it's when the edge actually represents two different interactions.

For example, Gene X - synthetic lethal - Gene Y Gene X - affinity capture western - Gene Y

In this case synthetic lethal is genetic and affinity capture is physical, but I think that it shows up as a yellow line. I will try to find the example I was looking at yesterday. It's possible there was both a yellow line and a green line but I could only see the yellow?

Christie

On Jan 6, 2012, at 11:33 AM, Sven Heinickereply@reply.github.com wrote:

Disregard my previous message, but according to the TAB2 docs a single row can never be both.

http://wiki.thebiogrid.org/doku.php/biogrid_tab_version_2.0

column 13 says: Experimental System Type. This will be either “physical” or “genetic” as a classification of the Experimental System Name.

Unlike row 18 where items can be High Throughput, Low Throughput or Both (separated by “|”)

can you give me an example edge that is both?

---

Reply to this email directly or view it on GitHub: https://github.com/svenmh/BioGRID-Viewer/issues/13#issuecomment-3389524

[svenmh]

Can you send be biogrid edge ids? Or the two gene that the edges connect. Or even the SVG URL.

[RoseO]

Hi Sven, I now see the problem and I'm not sure you can do much about this for
the first release. Your program is displaying both genetic and
physical interactions in their respective edge colors as expected,
but they're looping away from each other so there can't be an
overlapping blue color for both physical and genetic interactions.
Also, it's possible to not see one of them if they've looped far away
giving the impression that it's missing from the image. If you look
at FUS3, it has a number of interactions with both genetic and
physical evidence codes associated with them (http://thebiogrid.org/ 32682/summary/saccharomyces-cerevisiae/fus3.html). The thing is, in
the interactions file, they put all the evidence codes on separate
lines, even for the same interactors, and mark them as either
physical or genetic, but not both. Your program would have to merge
that information. So here you can see that FUS3 interacts with DIG1
with different evidence codes on different lines (Biogrid IDs
113363, 151835, 156727, 156728):

113363 856058 852091 36131 32532 YPL049C YDR480W DIG1 DIG2 RST1 RST2
Two-hybrid physical Uetz P (2000) 10688190 4932 4932 High
Throughput - - - - - BioGRID|INTACT|MINT 151835 852265 856058 32682 36131 YBL016W YPL049C FUS3 DIG1 DAC2 RST1
Biochemical Activity physical Tedford K (1997) 9094309 4932 4932 Low
Throughput - Phosphorylation - - - BioGRID

156727 856058 852091 36131 32532 YPL049C YDR480W DIG1 DIG2 RST1 RST2
Phenotypic Enhancement genetic Breitkreutz A (2003) 14668360 4932
4932 Low Throughput - - invasive growth|cell shape - - BioGRID 156728 856058 852091 36131 32532 YPL049C YDR480W DIG1 DIG2 RST1 RST2
Phenotypic Enhancement genetic Gelli A (2002) 12410840 4932 4932 Low
Throughput - - mating response - - BioGRID

Your program draws the lines accordingly and also weights them
according to the number of genetic or physical interactions (thicker
lines). Maybe in the second release there could be an option to see
genetic (green) or physical (yellow) or both (blue), but I'm not sure
this can easily be done for the first version. I think it might work
as an option in the standalone version. What do you think? Thanks, Rose

On Jan 6, 2012, at 2:33 PM, Sven Heinicke wrote:

Disregard my previous message, but according to the TAB2 docs a
single row can never be both.

http://wiki.thebiogrid.org/doku.php/biogrid_tab_version_2.0

column 13 says: Experimental System Type. This will be either
physical or genetic as a classification of the Experimental
System Name.

Unlike row 18 where items can be High Throughput, Low Throughput
or Both (separated by |)

can you give me an example edge that is both?

---

Reply to this email directly or view it on GitHub: https://github.com/svenmh/BioGRID-Viewer/issues/ 13#issuecomment-3389524

[RoseO]

cc'ing Christie because I'm not sure who can see these github emails.

On Jan 6, 2012, at 3:53 PM, Rose Oughtred wrote:

Hi Sven, I now see the problem and I'm not sure you can do much about this
for the first release. Your program is displaying both genetic and
physical interactions in their respective edge colors as expected,
but they're looping away from each other so there can't be an
overlapping blue color for both physical and genetic
interactions. Also, it's possible to not see one of them if
they've looped far away giving the impression that it's missing
from the image. If you look at FUS3, it has a number of
interactions with both genetic and physical evidence codes
associated with them (http://thebiogrid.org/32682/summary/ saccharomyces-cerevisiae/fus3.html). The thing is, in the
interactions file, they put all the evidence codes on separate
lines, even for the same interactors, and mark them as either
physical or genetic, but not both. Your program would have to
merge that information. So here you can see that FUS3 interacts
with DIG1 with different evidence codes on different lines
(Biogrid IDs 113363, 151835, 156727, 156728):

113363 856058 852091 36131 32532 YPL049C YDR480W DIG1 DIG2 RST1
RST2 Two-hybrid physical Uetz P (2000) 10688190 4932 4932 High
Throughput - - - - - BioGRID|INTACT|MINT 151835 852265 856058 32682 36131 YBL016W YPL049C FUS3 DIG1 DAC2
RST1 Biochemical Activity physical Tedford K (1997) 9094309 4932
4932 Low Throughput - Phosphorylation - - - BioGRID

156727 856058 852091 36131 32532 YPL049C YDR480W DIG1 DIG2 RST1
RST2 Phenotypic Enhancement genetic Breitkreutz A (2003) 14668360
4932 4932 Low Throughput - - invasive growth|cell shape - - BioGRID 156728 856058 852091 36131 32532 YPL049C YDR480W DIG1 DIG2 RST1
RST2 Phenotypic Enhancement genetic Gelli A (2002) 12410840 4932
4932 Low Throughput - - mating response - - BioGRID

Your program draws the lines accordingly and also weights them
according to the number of genetic or physical interactions
(thicker lines). Maybe in the second release there could be an
option to see genetic (green) or physical (yellow) or both (blue),
but I'm not sure this can easily be done for the first version. I
think it might work as an option in the standalone version. What
do you think? Thanks, Rose

On Jan 6, 2012, at 2:33 PM, Sven Heinicke wrote:

Disregard my previous message, but according to the TAB2 docs a
single row can never be both.

http://wiki.thebiogrid.org/doku.php/biogrid_tab_version_2.0

column 13 says: Experimental System Type. This will be either
physical or genetic as a classification of the Experimental
System Name.

Unlike row 18 where items can be High Throughput, Low Throughput
or Both (separated by |)

can you give me an example edge that is both?

---

Reply to this email directly or view it on GitHub: https://github.com/svenmh/BioGRID-Viewer/issues/ 13#issuecomment-3389524

[ChristieChang]

Yes, Rose's example is what I was referring to. I only realized it when I was looking at a single publication (by Pubmed ID), and tried to look at only "genetic interactions" and thought it was weird that some of them were yellow. I then realized that they represented both physical and genetic interactions. I think Rose's suggestions for making it part of the second release would be fine.

[svenmh]

I'm still not getting it, but if you send me the pubmed ID and I select only genetic interactions it shows phy&gen interactions? Please pass on the pubmed id.

(If I still don't get it after that lets share a terminal using our meeting software and you show me)

[RoseO]

Sorry for not being clear. Whether they're in the same pmids or
different ones, any interactions between the same two interactors
(e.g. fus3 + dig1) can be either genetic or physical. We were trying
to say that if between two interactors there are both physical and
genetic interactions, it might be worth making single blue edges
(replacing the separate green and yellow edges), instead of looping
out separate green and yellow edges. However, your program can't
directly get a "both" genetic and physical interaction edge directly
from the interactions file because there's no "both"option in the
physical/genetic field, as you pointed out. It would require your
program to re-label the interaction as "both" if there were both
genetic and physical interactions between the same too interactors.
However, this probably has to be run past the whole group because
there are good reasons not to have this. For example, if there are
more genetic than physical interactions and the genetic line is a
thicker edge, that information would be lost if it were combined into
a single blue edge. In that case, the user couldn't tell if there
was more genetic or physical evidence for an interaction. However,
there could be an option for users in the standalone version to have
either the combined blue edge for both phys/gen interactions between
interactors, or else separate green/yellow edges. All of this to
say, please don't worry about this request for the first release.
Hope this is a bit more clear. :)

On Jan 9, 2012, at 1:54 PM, Sven Heinicke wrote:

I'm still not getting it, but if you send me the pubmed ID and I
select only genetic interactions it shows phy&gen interactions?
Please pass on the pubmed id.

(If I still don't get it after that lets share a terminal using our
meeting software and you show me)

---

Reply to this email directly or view it on GitHub: https://github.com/svenmh/BioGRID-Viewer/issues/ 13#issuecomment-3418093

[svenmh]

Are you saying that in a perfect world the interaction between two genes can only physical or genetic? If one paper say it is physical and the other says it is genetic then one of the papers is wrong?

[RoseO]

I tried calling you, but I think you're working from home today? Can
you please send me your phone number? I can explain this better and
I have a few other questions too. Thanks!

On Jan 9, 2012, at 2:56 PM, Sven Heinicke wrote:

Are you saying that in a perfect world the interaction between two
genes can only physical or genetic? If one paper say it is physical
and the other says it is genetic then one of the papers is wrong?

---

Reply to this email directly or view it on GitHub: https://github.com/svenmh/BioGRID-Viewer/issues/ 13#issuecomment-3419125

[svenmh]

I've added a colorize option on the bundle view, so you can turn on of of colors.