Network of MIMP predicted kinases

[krassowski]

We may add an additional "network view" to show novel interactions predicted by MIMP (currently we only show losses in network view; gains are accessible only in sequence view).

[krassowski]

@reimand0 this is mostly complete: https://activedriverdb.org/network/predicted/NM_007294.

I added families so we show all information predicted with MIMP. If there are too many nodes on some networks, we could group kinases by families (i.e. hide kinases which belong to families until a family is tapped on) or hide the family (group) nodes completely.

Please, let me know of any comments and requests.

[krassowski]

I used green color and rounded dashed line (for predicted gains). There are more losses shown in predicted network than in the normal one as in the normal there are shown only kinases which are both: known to interact with given site and with predicted loss.

[reimand0]

this looks great, thanks! There is one strange feature in the visualisation - the network appears to be duplicated with every node represented twice.

[reimand0]

let's update the titles in the tabs network view, experimental network view, MIMP-predicted

[krassowski]

In which browsers did you get the duplicated nodes?

6 sie 2017 1:26 PM "reimand0" notifications@github.com napisał(a):

let's update the titles in the tabs network view, experimental network view, MIMP-predicted

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/reimandlab/ActiveDriverDB/issues/121#issuecomment-320501052, or mute the thread https://github.com/notifications/unsubscribe-auth/AFkAxjmBtDP768mF7orOPbNNAFkPGnYWks5sVaMBgaJpZM4Oohk6 .

[reimand0]

chrome on mac

[reimand0]

safari does not have that problem

[krassowski]

I changed the tab names.

I tested the network with Chrome 57, 58, 59, 60 and 61 developer on Sierra and I cannot reproduce the problem with nodes duplication. Also tested Chrome 60 on El Capitan, Yosemite and Mavericks.