BioNet-XR: Biological Network Visualization Framework for Virtual Reality and Mixed Reality Environments

TL;DR

BioNet-XR is a novel 3D visualization framework enabling immersive exploration of protein-protein interaction networks in virtual and mixed reality environments, enhancing collaborative analysis and usability.

Contribution

It introduces a modular, Unity3D-based XR framework with advanced visualization features and network algorithms, tailored for biological network exploration in VR and MR.

Findings

Usability tests with bioinformaticians and multidisciplinary groups.

Seamless transition between virtual and real environments in MR.

Framework adaptable for other biological networks.

Abstract

Protein-protein interaction networks (PPIN) enable the study of cellular processes in organisms. Visualizing PPINs in extended reality (XR), including virtual reality (VR) and mixed reality (MR), is crucial for exploring subnetworks, evaluating protein positions, and collaboratively analyzing and discussing on networks with the help of recent technological advancements. Here, we present BioNet-XR, a 3D visualization framework, to visualize PPINs in VR and MR environments. BioNet-XR was developed with the Unity3D game engine. Our framework provides state-of-the-art methods and visualization features including teleportation between nodes, general and first-person view to explore the network, subnetwork construction via PageRank, Steiner tree, and all-pair shortest path algorithms for a given set of initial nodes. We used usability tests to gather feedback from both specialists (bioinformaticians) and generalists (multidisciplinary groups), addressing the need for usability evaluations of visualization tools. In the MR version of BioNet-XR, users can seamlessly transition to real-world environments and interact with protein interaction networks. BioNet-XR is highly modular and adaptable for visualization of other biological networks, such as metabolic and regulatory networks, and extension with additional network methods.