Confidence scores differ between networks & STRING excludes some edges above confidence threshold

[alin...@ucr.edu]

Hello --

I'm working with STRINGApp v 1.4.0 implemented in cytoscape 3.6.0 on a Mac. I have sets of Drosophila melanogaster genes and am building networks at confidence level 0.6, with no additional interactors. I noticed inconsistencies in the connections between proteins in different networks and am trying to sort out why. Lists of gene names were imported via (import > network > public databases > STRING:protein query). Here is what I noticed about genes A,B,C in Networks #1 and #2.

Network #1 has ~250 genes, genes that are diferentially expressed under condition X
- gene A connects to gene B at confidence score 0.607
- gene B connects to gene C at confidence score 0.961
- The lowest scoring edge in the whole network is 0.6000001 -- threshold is behaving

Network #2 has ~450 genes, genes that are deferentially expressed under condition Y
- gene A does not connect to gene B when I set the confidence threshold to 0.6
- gene B connects to gene C at at confidence score 0.961, same as Network #1
- The lowest scoring edge in the whole network is 0.616 -- made me worry, as Network #1 had many edges just above 0.6.

In Network #2, I change the confidence to 0.5, now new edges have been added! But, when I sort the edge table, the lowest score of all edges is still 0.616, and the gene A to gene B edge score has now jumped up to 0.830, and gene B to gene C now has a score of 0.989. Change the confidence threshold back up to 0.6, and no edges are removed, scores stay high.

I played with the scores in Network #1 and nothing like this happened. I re-loaded in Network #2, and now only edges with score >0.805 are included in the network, even though threshold is at 0.6.

I can't seem to find any discussion of something similar happening. At first I was worried that the score of an edge was being scaled concomitant with total node number, but that does not seem to be the case at all, given the inconsistencies within a network, not just between networks.

Any advice is appreciated.

Many thanks,
Amelia

[Scooter Morris]

Hi Amelia,

   Can you send me your two gene lists view private mail?  I'll try to reproduce and then figure out what's going on.  You are correct that this is not the expected behavior.

-- scooter

[A Lindsey]

Hi! I sent you the gene lists -- hopefully they went though; let me know if you did not get them. Looking forward to figuring this out!

Thanks,

Amelia

[Scooter Morris]

Hi Amelia,

  No, they never came through.  Can you try again and let me know what the subject line will be?

-- scooter

[A Lindsey]

Yep -- just sent them again, subject line "Gene lists for Cytoscape/STRING troubleshooting"

Thanks!

A

[Scooter Morris]

Hmm.... OK, still didn't get them :-(  Let's do it this way -- go to https://github.com/RBVI/stringApp/issues and report this as a bug.  You can attach the gene list there.

-- scooter

[Alexander Pico]

From: Amelia Lindsey <alin...@ucr.edu>

Subject: Re: [cytoscape-helpdesk] Re: Confidence scores differ between networks & STRING excludes some edges above confidence threshold

Date: January 4, 2019 at 12:29:00 PM PST

To: cytoscape...@googlegroups.com

Done. https://github.com/RBVI/stringApp/issues/13

Amelia Lindsey
Postdoctoral Scholar
Department of Biology
Indiana University
Newton Lab, Jordan Hall 221
1001 East Third Street
Bloomington, IN 47405
Email: aml...@iu.edu
Twitter: @Amelia_Lindsey
matryoshkasymbiosis.wordpress.com

[Scooter Morris]

Hi Amelia,

   Can you give stringApp 1.4.1 a try?

-- scooter

[Amelia Lindsey]

Okay the new version seems to be better: the threshold is behaving and the edge scores do not shift when I change the confidence level.

The score of an edge between two unique nodes is the same regardless of which network it is in. The minor inconvenience is that the scores are now different than when I was using STRING 1.4.0 — was this expected? The A-B gene connection which I mention previously and consistently was at a score 0.607 in network #1, is now consistently a score of 0.567 in both networks #1 and #2. If this is expected, then I need to make some decisions on how to move forward RE: finalizing networks for publication. :)  

Thanks!

Amelia

Amelia Lindsey
Postdoctoral Scholar
Department of Biology
Indiana University
Newton Lab, Jordan Hall 221
1001 East Third Street
Bloomington, IN 47405
Email: aml...@iu.edu
Twitter: @Amelia_Lindsey
matryoshkasymbiosis.wordpress.com

--
You received this message because you are subscribed to a topic in the Google Groups "cytoscape-helpdesk" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/cytoscape-helpdesk/eq1mlK3a-U0/unsubscribe.
To unsubscribe from this group and all its topics, send an email to cytoscape-helpd...@googlegroups.com.
To post to this group, send email to cytoscape...@googlegroups.com.
Visit this group at https://groups.google.com/group/cytoscape-helpdesk.
To view this discussion on the web visit https://groups.google.com/d/msgid/cytoscape-helpdesk/2d2fe0da-59b7-4c29-b9a9-f0e9559d93b0%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Scooter Morris]

On Friday, January 11, 2019 at 11:14:45 AM UTC-8, Amelia Lindsey wrote:

Okay the new version seems to be better: the threshold is behaving and the edge scores do not shift when I change the confidence level.

The score of an edge between two unique nodes is the same regardless of which network it is in. The minor inconvenience is that the scores are now different than when I was using STRING 1.4.0 — was this expected?

 

Yes,  it is to be expected.  There was a slight difference in the way the app calculated the score and the way the website calculated the score.  We now just download the score directly from the website.

The A-B gene connection which I mention previously and consistently was at a score 0.607 in network #1, is now consistently a score of 0.567 in both networks #1 and #2. If this is expected, then I need to make some decisions on how to move forward RE: finalizing networks for publication. :)  

-- scooter

Thanks!

Amelia

To unsubscribe from this group and all its topics, send an email to cytoscape-helpdesk+unsub...@googlegroups.com.