The HRA Organ Gallery is a VR application that enables users to explore 3D organ models of the HRA in their true size, location, and spatial relation to each other. The HRA Organ Gallery has two main use cases: 1) introducing both novice and expert users to the HuBMAP data available via the HuBMAP Data Portal, and 2) providing quality assurance and quality control (QA/QC) for HRA data providers. Additional use cases include: education and outreach, enhancing patient education, optimizing data analysis in the field of histology; more use cases are under development. Further information can be found in our recent Frontiers in Bioinformatics paper. The application is available for the Meta Quest 2, 3, and Pro on the Meta Horizon Store, free of charge.

The HRA, led by the Human BioMolecular Atlas Program (HuBMAP), aims to map the adult healthy human body at single-cell resolution through a collaboration across 20+ international consortia. The project includes three main categories of data: biological structure, spatial, and specimen data. The ASCT+B tables are compiled by experts to capture biological structure data, describing the connection between anatomical structures, cell types, and biomarkers. As of October 7, 2024, the HRA Organ Gallery features 65 3D reference organs, 906 mapped tissue blocks from 370 demographically diverse donors and 27 providers that link to 6,000+ datasets; it also features prototype visualizations of single-cell data in 3D with immersive itneraction abilities, as well as an elevator system to traverse the human body across scales.

Alex Wong, Ph.D.

Bio: I’m a postdoctoral fellow in the Laboratory of Systems Pharmacology at Harvard Medical School, supervised by Prof. Peter Sorger. My current research interest is to create high-resolution maps to better understand the 3D spatial relationships of immune cell markers, cell states and nerves in the same millimeter thick cleared intact human FFPE colon tissue. This 3D multiplexed imaging will be achieved through his newly developed cyclic method for cleared tissue with light-sheet microscopy. I obtained my Ph.D. in Bioengineering from Karolinska Institutet from the Ming Wai Lau Centre for Reparative Medicine (Hong Kong) and Dept. of Neuroscience. Supervised by Prof. Sijie Chen, Prof. Ola Hermansson & Prof. Per Uhlen. My work involved developing novel 3D imaging techniques with Aggregation-Induced Emission based fluorescent dyes in cleared tissue.

Abstract: Advancements in data collection and computational methods are essential for fully describing and quantifying the intricate organization and biology of healthy and diseased tissues. Understanding key features such as immune-cell interactions, epithelial cell morphology and state, and macroscale structures like supporting vessels and nerves requires comprehensive analysis at multiple spatial scales. These complexities necessitate innovative approaches to data acquisition, processing, and visualization to unravel the underlying biological mechanisms and disease processes. In my current research, I am collecting and analyzing a fundamentally new type of high-plex three-dimensional (3D) data from normal and diseased colon tissues. By referencing both the Human Tumor Atlas Network (HTAN) and the Human BioMolecular Atlas Program (HuBMAP) ontologies, I aim to create a detailed map of cellular interactions and tissue architecture. The samples I work with are up to several millimeters thick and can be reliably imaged across many channels using advanced multiplexed light-sheet microscopy techniques. This allows for the capture of different spatial scales while maintaining single-cell resolution in each acquisition cycle. However, the sheer volume of data presents significant challenges. Each dataset, composed of multiple imaging cycles, can accumulate tens of terabytes. This massive data size makes it difficult to quickly inspect and analyze the information using traditional visualization and analysis tools. Rapid and efficient data inspection is crucial for identifying patterns, anomalies, and interactions that could lead to new insights into tissue biology and disease mechanisms. I am eager to attend the Powers of Ten workshop because it focuses on developing new virtual reality (VR) visualization methods and interfaces which will be essential for the comprehensive exploration of such large and complex multiplexed light-sheet datasets. The workshop provides a unique opportunity to collaborate with experts in VR visualization, data encoding, and biological data analysis. In particular, I am interested in exploring how data can be encoded into alternative representations, such as mesh models, which can significantly reduce storage space requirements while retaining all the necessary spatial information for analysis. My project specifically focuses on neuroimmune interactions within colon tissue. Understanding these interactions at both micro and macro scales is critical for elucidating the mechanisms of diseases such as colorectal cancer and inflammatory bowel disease. Furthermore, developing an “exploded view” mode in VR could greatly enhance our ability to highlight areas of interaction and inspect individual cells in detail. This functionality would allow researchers to deconstruct complex tissue architectures in a virtual environment, providing new perspectives and insights that are not possible with traditional two-dimensional (2D) imaging or even standard 3D visualization tools. From our experience with processing and visualizing 3D data, I recognize the potential of VR to revolutionize scientific visualization of multiscale datasets. By immersing researchers in a virtual environment, VR can facilitate a more intuitive understanding of spatial relationships and interactions within tissues. Collaborating with the Human Reference Atlas (HRA) VR working group at the workshop could help bridge the gap between data generation and meaningful interpretation, leading to the creation of an intuitive VR visualization platform for 3D spatial omics. Attending the Powers of Ten workshop will enable me to share my insights and challenges with a community of like-minded researchers and technologists. I hope to contribute my experience in high-resolution imaging and data processing to the development of innovative VR visualization tools. I also hope to learn from the expertise of others in the fields of VR interface design, data compression, and multiscale visualization techniques.

Yang Miao

Bio: I’m a Ph.D. student in the Department of Biomedical Engineering at Duke University, working under the supervision of Dr. John Hickey. My current research focuses on exploring spatial relationships among cells using multiplexed imaging. I obtained my M.S. in Biomedical Engineering at Johns Hopkins University and my B.S. in Biotechnology at Beijing Normal University in China.

Abstract: Working in a spatial proteomics lab, my research focuses on understanding the complex networks of cell interactions through multiplexed imaging techniques, such as CODEX. I am eager to attend the workshop to deepen my understanding of advanced biovisualization techniques and explore how these tools can be applied to the analysis of spatial-omics datasets. One of my key interests is learning how to effectively visualize the datasets in 3D. Integrating spatial data with current virtual reality (VR) technology offers a unique opportunity to explore organs and tissues at their corporal locations. It provides an interactive and user-friendly platform for analyzing intricate spatial datasets; I am particularly excited to learn how VR can be leveraged to uncover new biological insights. Understanding the current challenges in biovisualization is another area of focus for me. The hands-on sessions with VR devices are an exciting aspect of the workshop, as they will provide practical insights into how these technologies can be integrated into existing data analysis workflows. Incorporating these tools into regular research practices could significantly enhance the interpretation of spatial datasets, benefiting both researchers and clinicians alike. In addition, I look forward to sharing our expertise in visualizing the hierarchical organization of tissues, including cell types, tissue neighborhoods, and broader community structures during the workshop. Developing multiscale visualizations of these elements could yield novel perspectives on tissue architecture and spatial relationships, making the exploration of multiplexed imaging data more intuitive and impactful. Thus, I am especially interested in how to optimize the techniques to highlight multiscale cellular neighborhoods in 3D visualizations. A further goal is to connect with fellow Junior Investigators in the HuBMAP community, fostering a collaborative environment where ideas and feedback on visualization techniques can be exchanged. This interaction will help me better understand how others in the field are addressing similar challenges in spatial-omics research. By attending this workshop, I aim to gain insights into the latest advancements in 3D biovisualization, along with practical skills that can be directly applied to my work with HuBMAP data. In return, I hope to contribute our experiences with spatial multiplexed imaging and multiscale analysis while engaging in meaningful discussions about the future of data visualization in spatial omics.

Archibald Enninful

Bio: I’m a PhD student in Prof. Rong Fan’s group at Yale. I first joined the Fan Lab during my undergraduate studies, where I applied spatial omics technologies to profile the mouse lymph node microenvironment. Now, as a graduate student in the lab, my research focuses on mapping senescent cells in lymphoid tissues as part of the NIH’s Cellular Senescence Network (SenNet) Consortium. Additionally, I am part of the Human Tumor Atlas Network (HTAN), working to construct 3D atlases of human cancers, specifically lymphomas. Through my research, I aim to advance the understanding of aging and cellular behavior in complex tissues.

Abstract: Advances in spatial multiomics have transformed our ability to study tissue architecture, capturing details of cellular interactions and molecular states. However, the vast amount of data generated—often terabytes per sample—poses significant challenges for visualization and analysis. Virtual Reality (VR) offers an immersive platform to explore spatial multiomic datasets. Techniques like CODEX and spatial transcriptomics capture the spatial distribution of proteins and RNA within tissues at single-cell resolution. In my research with the NIH’s Cellular Senescence Network (SenNet) and the Human Tumor Atlas Network (HTAN), I focus on mapping the lymph node microenvironment and senescent cells. Senescence, a process linked to aging and cancer, requires spatial context to fully understand its impact on tissue behavior. Traditional 2D imaging often fails to capture these important dynamics, especially in complex tissues like lymph nodes and tumors. VR addresses this challenge by enabling 3D exploration of tissue architecture, providing new ways to visualize cell-cell interactions and tissue structure. By rendering features such as immune cells, epithelial structures, and vasculature, VR offers an intuitive platform for analysis. In my work, VR allows detailed examination of senescent cells within lymph nodes, revealing patterns that may be missed through 2D methods. Overall, VR is a powerful tool for visualizing spatial multiomic data, enabling immersive exploration of tissue structures and uncovering new insights into processes like senescence and cancer. I look forward to connecting with fellow Junior Investigators across NIH consortia to further explore the use of VR for spatial omics data visualization.

Chenchen Zhu, Ph.D.

Bio: Chenchen Zhu is currently a research scientist in Michael Snyder’s lab at Stanford University. He obtained a PhD in Genetics and computational biology in European Molecular Biology Laboratory. His current research focuses on dissecting the complexities of healthy and diseased intestine with single-cell RNA sequencing and spatial transcriptomics technologies. These efforts are instrumental in uncovering the subtle nuances of cell type composition and tissue architecture in both normal and pathological conditions of the human intestine and heart. Previously, Chenchen developed a full-length transcript sequencing method based on Oxford Nanopore technologies to identify mis-spliced transcripts in familial dilated cardiomyopathy.

Abstract: Mapping the precise architecture of human tissues in a preserved 3D environment is key in understanding normal tissue functions and how they are subverted in disease. Spatial transcriptomics provides deep insight into cellular states and programs and is increasingly capable of single-cell analysis, particularly when dissociative single-cell data is also available. Recently, we have developed an approach to reconstruct 3D tissue maps for the human gut from serial thin sections profiled with spatial transcriptomics or high-plex immune-based protein imaging. We have discovered that few if any cells in traditional 5 µm sections are complete, substantially complicating accurate phenotyping, Moreover, 2D images miss key features of tissue ultrastructure that are readily detected in 3D. These observations underscores the importance of 3D tissue mapping using highly multiplexed assays. Our approach using serial section reconstruction is the least “sophisticated” approach for assaying 3D molecular maps, but it has the great advantage of enabling multi-omic analysis, for example with one set of sections devoted to transcript profiling and an interleaved set to protein-based imaging. Some spatial transcriptomic methods are also compatible with subsequent protein-based tissue imaging. The key challenge is to reconstruct the 3D model from these serial images through image or feature based alignment. To this end, we have built a robust 3D-reconstruction method called Space-Map that aligns tissue sections using multimodal features and reconstructing detailed tissue maps. To appreciate the stereotyped organization of the normal colon, 3D visualization is crucial. My focus for the workshop is to collaborate with the team to find the best ways to visualize our 3D models at multiple levels—including tissues, cells, and molecules (RNA). These represent vastly different scales in biology and are essential for understanding tissue biology. Using 3D rendering in virtual reality will be key to delivering the stereoscopic sense of the models, helping us improve our 3D reconstruction approach. To assess the 3D model, I will specially focus on stem cells and their descendants in the colonic villus that are involved in the normal function of the intestinal mucosa. The mucosa is also rich in immune cells, some of which are distributed through the tissue and others of which are concentrated in secondary lymphoid structures in Peyer’s patches. We will pay particular attention to the analysis of these diverse tissue-resident immune cells with the goal of more precise and accurate cell-type calling. In summary, this VR workshop will connect me with other 3D map builders working on various tissues within HuBMAP. By utilizing the power of VR, I aim to assess our 3D models and uncover new biological discoveries that are not readily visible with 2D rendering. Additionally, I am personally very interested in learning and understanding VR development and hope to apply this technology to other genomic visualizations.

Andreas ‘Andi” Bueckle, Ph.D.

Bio: I am the Research Lead in the Cyberinfrastructure for Network Science Center (CNS) at the Luddy School of Informatics, Computing, and Engineering at Indiana University Bloomington (and affiliated with the Luddy Data Science Program). My research interest is interactive information visualization in virtual reality (VR), augmented reality (AR), and other immersive techniques. Born and raised in Germany, I hold a B.A. in Media Studies from Eberhard Karls University in Tuebingen, an M.A. in Communications from Berlin University of the Arts, and a Ph.D. in Information Science from Indiana University. From early on, I developed a deep interest in digital artifacts, most notably videography and photography. After working as a video journalist and cameraman on projects in Germany, France, India, and the US, I decided to switch to a more technical education and started to pursue and finish my Ph.D. in Information Science, working with Dr. Katy Börner at Indiana University Bloomington.

Abstract: I am the lead developer of the HRA Organ Gallery in virtual reality (VR), a virtual reality (VR) application that enables users to explore 3D organ models of the HRA in their true size, location, and spatial relation to each other (app, paper). The HRA Organ Gallery has three main use cases: 1) introducing both novice and expert users to the 2D and 3D data available in the HRA via the HuBMAP Data Portal, the SenNet Data Portal, and similar efforts, and 2) providing quality assurance and quality control (QA/QC) for HRA data providers, and (3) as a tool to support data outreach for HuBMAP and other consortia. More use cases are under development. I am the recipient of a NIH HuBMAP JumpStart Fellowship to organize this workshop.

To come after the event!