Networks become (virtual) reality

Networks are a powerful tool to understand complex systems. We built VRNetzer, a virtual reality platform to interactively explore, visualize, and analyze networks. We leverage the fully immersive VR environment to explore genome-scale networks and identify the cause of a rare genetic disease.
Published in Protocols & Methods
Networks become (virtual) reality
Like

Data, data, data.  It was predicted that almost 60 zettabytes of data were created, captured, copied or consumed in the last year, almost an increase of 5000% from 2010(Rydning 2017).  As network scientists, we are most interested in the relationships between the data, and how different components in a complex system interact with each other.  There’s a problem, though.  Most interesting networks are simply too large and too complex to be visualized in a meaningful way -- most of the time we only see hairballs.

This is the network of interactions between proteins in human cells.  It is comprised of 16376 proteins and 309355 interactions.  In two dimensions, we can't understand anything!  All we see is a giant grey blob.

What if we could blow up this network to the size of a cathedral and stand in the middle of it?  And zoom into specific parts to understand the details, or zoom out for a more global bird’s-eye view?  Our goal is to connect the intuitive visual nature of networks with the powerful computational analysis tools of network theory.  With virtual reality this is possible -- in our immersive VR platform, we can explore and navigate seamlessly through networks, with full six degrees of freedom and infinite scaling of network size. 

Christiane V.R. Hütter is using the VRNetzer to explore specific genes.  She's selected a DNA Cross-Link repair gene that is often a culprit in SCID (Severe Combined Immunodeficiency Disease).  The panel on the right displays information about the gene: a brief description, the gene symbol, disease/phenotype associations, gene ontology annotations, and tissue specific expression levels.  The left panel shows the immediate disease neighborhood.
Felix Müller is using the VRNetzer to explore disease neighborhoods.  This layout emphasizes the similarity between different genes according to their disease associations.  Two genes that are close together are involved in a similar set of diseases.  In the popup window, Felix is reading the abstracts of relevant articles for the selected gene.

Twenty years ago, this was unimaginable -- but with the advancements in gaming technology and computing power, huge networks become (virtual) reality.  Our platform is built in an open and modular fashion: the core of our platform, the VR module, is implemented on Unreal engine 4, the industry gold standard in the video game world.  It boasts astounding graphics performance, maximal compatibility with a variety of VR headsets, and it’s open source, too! The analytics module is written in Python and served with Flask.  It contains all the code for network visualization and analysis, as well as helper functionality to upload new networks into the tool and download them to a format compatible with other network analysis tools.  The two modules are linked together with a user interface module coded in Python/Flask, Javascript and html.  All of these modules are open source and can be further customized by the user.

The VRNetzer platform consists of 5 modules that can be customized independently.  The frontend consists of the VR interface and an additional web browser-based interface to facilitate seamless exchange between VR sessions and standard working environments. The user interface module connects the VR module to the backend, which consists of two separate modules for data storage and data analytics.

In the paper we showcase how we use VRNetzer to help identify potential causes of rare genetic diseases.  Every 4 seconds a baby is born with a rare genetic disease.  Globally, rare diseases affect 1 in 20 people.  These diseases can be life threatening and most of the time they are caused by one specific gene mutation.  The question is… which gene?  The human genome project gave us the catalog of all the parts in the system, and the networks we build describe the wiring diagram between the parts -- this is where biological function emerges.

In one case, we had a patient suffering from severe combined immune deficiency (SCID) with an unknown genetic cause.  The patient presented with recurrent severe infections, fever, and thrush.  An immune panel revealed a low T-cell count and defective T-cell activation.  Next generation sequencing of the patient’s DNA revealed several hundred variants; by applying filters for frequency and quality, confirmation by Sanger sequencing, and comparisons with the patient’s pedigree (segregation analysis), we are left with 15 candidate genes.  By examining the local and global network neighborhoods of these genes in several functional contexts, we can collect the pieces of the puzzle to find the cause of the patient's symptoms.

This is but a very specific use case for our platform -- because you can plug in your own networks and your own code, our platform can be used to explore all kinds of networks and answer a wide range of questions.  Add your own and send us a screen capture!  We'd love to see your networks.  Currently VRNetzer ships with the network of airports as well as the human interactome.

Grab your VR headset and try it out!  Instructions on how to set it up can be found here: https://github.com/menchelab/VRNetzer/blob/main/README.md.  Don’t have a headset?  Well, we also have a 2D version you can run from your computer -- but trust us, the VR version is much more fun!

Happy VRnetzing!

This endeavor was made possible through our super cool interdisciplinary team: an architect, 2 artists, a biologist, a medical doctor, 2 physicists, a software engineer, and a systems biologist!  We had a blast building this together and we hope you’ll enjoy it too!

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in

Follow the Topic

Biological Techniques
Life Sciences > Biological Sciences > Biological Techniques

Related Collections

With collections, you can get published faster and increase your visibility.

Biology of rare genetic disorders

This cross-journal Collection between Nature Communications, Communications Biology, npj Genomic Medicine and Scientific Reports brings together research articles that provide new insights into the biology of rare genetic disorders, also known as Mendelian or monogenic disorders.

Publishing Model: Open Access

Deadline: Oct 30, 2024

Advances in catalytic hydrogen evolution

This collection encourages submissions related to hydrogen evolution catalysis, particularly where hydrogen gas is the primary product. This is a cross-journal partnership between the Energy Materials team at Nature Communications with Communications Chemistry, Communications Engineering, Communications Materials, and Scientific Reports. We seek studies covering a range of perspectives including materials design & development, catalytic performance, or underlying mechanistic understanding. Other works focused on potential applications and large-scale demonstration of hydrogen evolution are also welcome.

Publishing Model: Open Access

Deadline: Dec 31, 2024