Plugin for Unreal

[dborgesr]

Hey,
First wanted to say, awesome work.
I'm attempting to use your library for the visualization of large scale biological graphical structures, namely looping DNA to individual regulatory elements, I can get all the JSONs loaded into Unreal using VARest, this is where i get a bit stuck. Would it be at all possible to package your tool as an Unreal Plugin?

I'm going to do some research on how to make that happen, but in the meantime its out of my comfort range.

--Thanks

[kwonoh]

Hi,

Thank you for your interest in this project.

You can visualize your data by following the description (Data preprocessing) in the README.md.
Also, this project does not use VARest for loading JSON. See /Source/ImsvGraphVis/IGVData.cpp for more details about data loading.

At this moment, I don't have a plan to make this project as an Unreal Engine plugin.
If you want to implement other things, it will be much easier to use this project as the starting point.

Please feel free to ask me any questions about the code and resources of this project.

Oh-Hyun

[dborgesr]

Ah i see, that's unfortunate, due to the complexity of the networks i'm using and the clustering methodology i won't be able to use your code as it stands, namely its dependence on hierarchical clustering would mangle my data far too much to be practical.
As to the VARest, i know you guys dont use it, I do in my project, what that allows is for logic to be baked into the visualization inside Unreal Blueprints, you dont have to pull, build, and render all networks, you can pull and render what, for example, you stare at in VR.
I'll try and learn some of the plugin writing stuff to get this going, it would increase the penetration and usage of it a TON, especially in the biology community as the state of our graph visualization is hilariously limited, Graphviz and Cytoscape are probably the best :(

[kwonoh]

Hi @dborgesr,

I am sorry for the late reply.

Ah i see, that's unfortunate, due to the complexity of the networks i'm using and the clustering methodology i won't be able to use your code as it stands, namely its dependence on hierarchical clustering would mangle my data far too much to be practical.

Current layout method, which is described in our paper, relies on a hierarchical clustering since it uses a variant of hierarchical edge bundling technique.
Thus, if you don't want to use a hierarchical clustering, our technique might not the one for you.

As to the VARest, i know you guys dont use it, I do in my project, what that allows is for logic to be baked into the visualization inside Unreal Blueprints, you dont have to pull, build, and render all networks, you can pull and render what, for example, you stare at in VR.

I don't understand what you mean by "you dont have to pull, build, and render all networks". Could you please elaborate more?

I'll try and learn some of the plugin writing stuff to get this going, it would increase the penetration and usage of it a TON, especially in the biology community as the state of our graph visualization is hilariously limited, Graphviz and Cytoscape are probably the best :(

We hear this a lot. Since the requirement of each project varies a lot, there are no good general purpose graph visualization software that works for all project. Generally, our research group develops different software per project.

Thank you,
Oh-Hyun